summary.txt

""
Series: . 
Regression with ARIMA(0,0,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag1_NPGO  lag1_PDO
       -1.0238        -1.5218     1.4153   -2.8888
s.e.    0.5163         0.0716     0.4372    0.6143

sigma^2 = 0.7881:  log likelihood = -7.63
AIC=25.25   AICc=55.25   BIC=25.65

Training set error measures:
                    ME   RMSE    MAE       MPE     MAPE      MASE      ACF1
Training set -42859.55 156136 100486 -18.72663 57.28275 0.8755411 0.2842426
Series: . 
Regression with ARIMA(0,0,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      intercept  pink_ind  lag1_log_SAR1  lag1_NPGO  lag1_PDO
        11.6464   -1.2491        -0.0374     0.9775   -0.1626
s.e.     0.8956    0.1403         0.1191     0.1304    0.2402

sigma^2 = 0.07143:  log likelihood = 2.75
AIC=6.49   AICc=48.49   BIC=7.67

Training set error measures:
                   ME     RMSE     MAE       MPE    MAPE      MASE       ACF1
Training set 5834.284 45947.51 32971.7 -1.559025 15.2309 0.3350242 -0.1434357
Series: . 
Regression with ARIMA(0,0,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      intercept  pink_ind  lag1_log_SAR1  lag1_NPGO  lag1_PDO
        13.6524    -1.321         0.2199     0.8344    0.0882
s.e.     1.2758     0.232         0.1719     0.2099    0.3881

sigma^2 = 0.176:  log likelihood = -2.04
AIC=16.08   AICc=44.08   BIC=17.89

Training set error measures:
                   ME     RMSE      MAE       MPE    MAPE      MASE      ACF1
Training set 12778.29 100669.3 58227.17 -4.382273 22.9187 0.5915794 0.1111684
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag1_NPGO  lag1_PDO
       -1.0962         0.4646     0.6596    0.9954
s.e.    0.1099         0.0834     0.1313    0.1764

sigma^2 = 0.1058:  log likelihood = -0.4
AIC=10.81   AICc=25.81   BIC=12.32

Training set error measures:
                   ME     RMSE      MAE     MPE   MAPE     MASE        ACF1
Training set 33577.74 68248.43 42992.35 6.33503 15.871 0.347768 -0.07659005
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag1_NPGO  lag1_PDO
       -1.0889         0.4798     0.6648    1.0097
s.e.    0.1049         0.0742     0.1262    0.1676

sigma^2 = 0.09275:  log likelihood = -0.04
AIC=10.08   AICc=22.08   BIC=12.07

Training set error measures:
                   ME     RMSE      MAE      MPE     MAPE      MASE        ACF1
Training set 26632.84 66893.17 41849.82 4.706968 14.90325 0.3023196 -0.09051352
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag1_NPGO  lag1_PDO
       -1.0892         0.4445     0.6950    0.9759
s.e.    0.1028         0.0551     0.1168    0.1579

sigma^2 = 0.08507:  log likelihood = 0.19
AIC=9.61   AICc=19.61   BIC=12.04

Training set error measures:
                   ME     RMSE      MAE     MPE     MAPE      MASE       ACF1
Training set 29943.05 69245.82 46249.23 5.64344 15.75625 0.2814921 -0.1759208
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag1_NPGO  lag1_PDO
       -1.0969         0.3507     0.5899    0.6403
s.e.    0.1642         0.0818     0.1829    0.2236

sigma^2 = 0.2089:  log likelihood = -5.88
AIC=21.75   AICc=30.33   BIC=24.58

Training set error measures:
                   ME     RMSE      MAE      MPE     MAPE      MASE       ACF1
Training set 38824.31 91742.31 64417.75 7.550373 24.10822 0.3952042 -0.1737504
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag1_NPGO  lag1_PDO
       -0.9242         0.3491     0.3940    0.4744
s.e.    0.1352         0.0868     0.1487    0.2125

sigma^2 = 0.2281:  log likelihood = -7.16
AIC=24.33   AICc=31.83   BIC=27.52

Training set error measures:
                   ME     RMSE      MAE      MPE     MAPE    MASE       ACF1
Training set 38358.68 91336.85 64528.96 8.392059 27.35055 0.42354 -0.1892946
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag1_NPGO  lag1_PDO
       -1.0003         0.3482     0.4287    0.5452
s.e.    0.1415         0.0948     0.1613    0.2287

sigma^2 = 0.2649:  log likelihood = -9
AIC=27.99   AICc=34.66   BIC=31.53

Training set error measures:
                   ME     RMSE      MAE      MPE     MAPE      MASE       ACF1
Training set 43599.62 97272.62 70826.53 11.42076 30.20799 0.4658427 -0.1850707
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag1_NPGO  lag1_PDO
       -0.9655         0.3414     0.4076    0.4595
s.e.    0.1337         0.0935     0.1577    0.2020

sigma^2 = 0.254:  log likelihood = -9.44
AIC=28.87   AICc=34.87   BIC=32.74

Training set error measures:
                   ME     RMSE      MAE      MPE    MAPE      MASE       ACF1
Training set 46009.28 99771.59 70739.28 12.03341 29.6134 0.4785562 -0.1745052
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag1_NPGO  lag1_PDO
       -0.9522         0.3383     0.3940    0.4607
s.e.    0.1210         0.0903     0.1454    0.1964

sigma^2 = 0.2355:  log likelihood = -9.55
AIC=29.1   AICc=34.56   BIC=33.27

Training set error measures:
                   ME     RMSE      MAE     MPE     MAPE      MASE       ACF1
Training set 39158.74 98305.37 71333.92 10.4957 28.79957 0.4495624 -0.1573529
Series: . 
Regression with ARIMA(0,0,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      intercept  pink_ind  lag1_log_SAR1
        16.2044   -0.7752         0.6263
s.e.     1.0327    0.3789         0.1561

sigma^2 = 0.459:  log likelihood = -6.36
AIC=20.71   AICc=34.05   BIC=21.03

Training set error measures:
                  ME     RMSE     MAE       MPE    MAPE      MASE       ACF1
Training set 14012.8 96206.91 69328.7 -14.20365 46.5496 0.6040655 -0.0813642
Series: . 
Regression with ARIMA(0,0,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      intercept  pink_ind  lag1_log_SAR1
        16.2997   -0.7014         0.6412
s.e.     0.9849    0.3467         0.1487

sigma^2 = 0.4005:  log likelihood = -6.83
AIC=21.66   AICc=31.66   BIC=22.44

Training set error measures:
                   ME     RMSE      MAE       MPE     MAPE      MASE       ACF1
Training set 11106.05 99139.97 68954.59 -13.80912 45.86001 0.7006451 -0.1451608
Series: . 
Regression with ARIMA(0,0,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      intercept  pink_ind  lag1_log_SAR1
        16.5049   -0.7579         0.6647
s.e.     0.8630    0.3157         0.1359

sigma^2 = 0.353:  log likelihood = -7.2
AIC=22.4   AICc=30.4   BIC=23.61

Training set error measures:
                   ME     RMSE      MAE       MPE     MAPE      MASE       ACF1
Training set 13181.53 103848.8 76525.86 -13.00673 44.54772 0.7774915 -0.1146319
Series: . 
Regression with ARIMA(0,0,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      intercept  pink_ind  lag1_log_SAR1
        16.7953   -0.5583         0.7120
s.e.     0.9780    0.3458         0.1539

sigma^2 = 0.4476:  log likelihood = -9.44
AIC=26.87   AICc=33.54   BIC=28.46

Training set error measures:
                   ME     RMSE      MAE      MPE     MAPE      MASE        ACF1
Training set 18311.89 158340.7 105502.7 -17.9476 49.96007 0.8534183 -0.02919362
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1
       -0.6427         0.3803
s.e.    0.1679         0.1251

sigma^2 = 0.3767:  log likelihood = -9.13
AIC=24.27   AICc=27.7   BIC=25.46

Training set error measures:
                   ME     RMSE      MAE       MPE     MAPE      MASE       ACF1
Training set 28565.43 131455.5 99884.97 -4.444521 39.84874 0.7215605 -0.1571202
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1
       -0.6282         0.3625
s.e.    0.1544         0.1057

sigma^2 = 0.3418:  log likelihood = -9.49
AIC=24.98   AICc=27.98   BIC=26.44

Training set error measures:
                   ME     RMSE      MAE       MPE     MAPE      MASE       ACF1
Training set 28959.58 124091.8 91733.66 -3.063598 37.52922 0.5583293 -0.1458682
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1
       -0.6709         0.3403
s.e.    0.1477         0.1030

sigma^2 = 0.335:  log likelihood = -10.25
AIC=26.5   AICc=29.17   BIC=28.2

Training set error measures:
                   ME     RMSE     MAE      MPE    MAPE      MASE       ACF1
Training set 29243.47 123440.4 91560.8 -1.18667 38.9498 0.5617273 -0.1598398
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1
       -0.6938         0.3444
s.e.    0.1389         0.1003

sigma^2 = 0.315:  log likelihood = -10.7
AIC=27.4   AICc=29.8   BIC=29.32

Training set error measures:
                   ME     RMSE      MAE       MPE     MAPE      MASE       ACF1
Training set 23982.93 121371.7 87974.55 -3.670279 39.34549 0.5774266 -0.1765039
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1
       -0.7646         0.3362
s.e.    0.1465         0.1093

sigma^2 = 0.3712:  log likelihood = -12.78
AIC=31.55   AICc=33.74   BIC=33.68

Training set error measures:
                   ME     RMSE      MAE        MPE     MAPE      MASE      ACF1
Training set 30086.16 135818.4 98701.77 -0.6272633 42.67967 0.6491848 -0.213176
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1
       -0.7740         0.3345
s.e.    0.1376         0.1059

sigma^2 = 0.3462:  log likelihood = -13.15
AIC=32.3   AICc=34.3   BIC=34.61

Training set error measures:
                   ME     RMSE      MAE       MPE     MAPE      MASE      ACF1
Training set 26083.58 132817.8 94570.42 -1.699886 41.31462 0.6397755 -0.248911
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1
       -0.7840         0.3436
s.e.    0.1302         0.0990

sigma^2 = 0.3251:  log likelihood = -13.51
AIC=33.01   AICc=34.86   BIC=35.51

Training set error measures:
                   ME     RMSE      MAE        MPE    MAPE     MASE       ACF1
Training set 29367.02 132481.4 95612.43 -0.6013718 40.1436 0.602571 -0.2591271
Series: . 
Regression with ARIMA(0,0,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      intercept  pink_ind  lag1_log_SAR1  lag1_NPGO
        12.0138   -1.2527         0.0031     1.0306
s.e.     0.6690    0.1485         0.1000     0.1378

sigma^2 = 0.07181:  log likelihood = 1.96
AIC=6.09   AICc=36.09   BIC=6.48

Training set error measures:
                   ME     RMSE      MAE       MPE     MAPE      MASE       ACF1
Training set 6843.904 55736.97 37514.45 -1.821366 15.69242 0.3268658 0.05847982
Series: . 
Regression with ARIMA(0,0,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      intercept  pink_ind  lag1_log_SAR1  lag1_NPGO
        12.0841   -1.2655         0.0133     1.0056
s.e.     0.6350    0.1417         0.0950     0.1267

sigma^2 = 0.06005:  log likelihood = 2.53
AIC=4.94   AICc=24.94   BIC=5.92

Training set error measures:
                   ME     RMSE      MAE       MPE     MAPE      MASE      ACF1
Training set 6346.251 52857.09 35542.85 -1.676765 15.20721 0.3611496 -0.033763
Series: . 
Regression with ARIMA(0,0,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      intercept  pink_ind  lag1_log_SAR1  lag1_NPGO
        13.4439   -1.3131         0.1963     0.8151
s.e.     0.8884    0.2299         0.1373     0.1925

sigma^2 = 0.1474:  log likelihood = -2.06
AIC=14.13   AICc=29.13   BIC=15.64

Training set error measures:
                   ME     RMSE      MAE       MPE     MAPE      MASE      ACF1
Training set 12503.19 100422.2 58215.27 -4.332322 23.19557 0.5914585 0.0770158
Series: . 
Regression with ARIMA(0,0,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      intercept  pink_ind  lag1_log_SAR1  lag1_NPGO
        16.2933   -0.6229         0.6357     0.1476
s.e.     1.3581    0.3630         0.2101     0.2806

sigma^2 = 0.499:  log likelihood = -9.3
AIC=28.6   AICc=40.6   BIC=30.59

Training set error measures:
                   ME     RMSE      MAE     MPE     MAPE      MASE       ACF1
Training set 19428.08 161078.7 106975.5 -16.818 47.24339 0.8653319 0.05557192
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag1_NPGO
       -0.8430         0.2671     0.3710
s.e.    0.2004         0.1353     0.2413

sigma^2 = 0.3488:  log likelihood = -8.06
AIC=24.13   AICc=30.79   BIC=25.72

Training set error measures:
                   ME     RMSE      MAE      MPE     MAPE      MASE        ACF1
Training set 49472.21 144484.6 97685.99 1.874766 32.24965 0.7056753 -0.04415366
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag1_NPGO
       -0.8310         0.3063     0.3152
s.e.    0.1922         0.1031     0.2032

sigma^2 = 0.3163:  log likelihood = -8.39
AIC=24.79   AICc=30.5   BIC=26.73

Training set error measures:
                   ME     RMSE      MAE        MPE     MAPE      MASE       ACF1
Training set 40243.11 136756.1 92962.02 -0.4004641 32.26848 0.5658056 0.01199235
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag1_NPGO
       -0.8775         0.2840     0.3275
s.e.    0.1855         0.1001     0.2022

sigma^2 = 0.3066:  log likelihood = -9.06
AIC=26.11   AICc=31.11   BIC=28.37

Training set error measures:
                   ME     RMSE      MAE      MPE     MAPE      MASE         ACF1
Training set 39971.86 134205.2 92009.06 1.445235 33.45382 0.5644774 -0.008074923
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag1_NPGO
       -0.8436         0.2886     0.2901
s.e.    0.1517         0.0960     0.1645

sigma^2 = 0.2812:  log likelihood = -9.3
AIC=26.59   AICc=31.04   BIC=29.15

Training set error measures:
                   ME     RMSE      MAE      MPE     MAPE      MASE        ACF1
Training set 37762.81 128075.8 86177.64 2.117139 32.39846 0.5656326 -0.02809266
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag1_NPGO
       -0.9223         0.2764     0.3137
s.e.    0.1616         0.1056     0.1807

sigma^2 = 0.3348:  log likelihood = -11.4
AIC=30.81   AICc=34.81   BIC=33.64

Training set error measures:
                   ME     RMSE      MAE      MPE     MAPE      MASE        ACF1
Training set 44248.71 137251.2 94921.48 5.350505 35.14422 0.6243209 -0.05536488
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag1_NPGO
       -0.9307         0.2747     0.3148
s.e.    0.1528         0.1022     0.1752

sigma^2 = 0.3102:  log likelihood = -11.68
AIC=31.36   AICc=34.99   BIC=34.45

Training set error measures:
                   ME   RMSE      MAE      MPE     MAPE      MASE        ACF1
Training set 39897.12 133542 90517.55 4.174232 33.92297 0.6123576 -0.09404484
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag1_NPGO
       -0.9149         0.2708     0.2984
s.e.    0.1380         0.0984     0.1605

sigma^2 = 0.2895:  log likelihood = -11.93
AIC=31.87   AICc=35.2   BIC=35.2

Training set error measures:
                   ME     RMSE      MAE      MPE     MAPE      MASE        ACF1
Training set 32764.99 130934.4 90220.31 2.853593 32.94356 0.5685886 -0.07042979
Series: . 
Regression with ARIMA(0,0,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag1_PDO
       -0.1464        -1.3741   -4.5223
s.e.    0.6678         0.0840    0.5325

sigma^2 = 1.456:  log likelihood = -10.98
AIC=29.95   AICc=43.28   BIC=30.27

Training set error measures:
                    ME     RMSE      MAE       MPE     MAPE     MASE      ACF1
Training set -46777.63 236744.4 152595.8 -39.87424 93.87731 1.329577 0.1070831
Series: . 
Regression with ARIMA(0,0,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      intercept  pink_ind  lag1_log_SAR1  lag1_PDO
        13.7844   -0.6983         0.3323    -0.736
s.e.     2.2857    0.3219         0.2920     0.613

sigma^2 = 0.4142:  log likelihood = -6.16
AIC=22.32   AICc=42.32   BIC=23.3

Training set error measures:
                   ME     RMSE      MAE       MPE     MAPE      MASE       ACF1
Training set 7112.047 84786.07 66200.71 -11.74011 44.26886 0.6726631 -0.2988268
Series: . 
Regression with ARIMA(0,0,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      intercept  pink_ind  lag1_log_SAR1  lag1_PDO
        14.7982   -0.7897         0.4540   -0.5356
s.e.     1.9961    0.3045         0.2594    0.5701

sigma^2 = 0.3784:  log likelihood = -6.78
AIC=23.55   AICc=38.55   BIC=25.07

Training set error measures:
                   ME     RMSE      MAE       MPE     MAPE     MASE       ACF1
Training set 11258.23 108302.3 82883.98 -11.84355 43.60676 0.842089 -0.2227417
Series: . 
Regression with ARIMA(0,0,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      intercept  pink_ind  lag1_log_SAR1  lag1_PDO
        17.9569   -0.6429         0.8491    0.4144
s.e.     1.2981    0.3302         0.1804    0.3288

sigma^2 = 0.447:  log likelihood = -8.69
AIC=27.39   AICc=39.39   BIC=29.38

Training set error measures:
                   ME     RMSE      MAE       MPE     MAPE      MASE        ACF1
Training set 16678.31 131785.5 94137.35 -15.86635 49.40886 0.7614833 -0.01187048
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag1_PDO
       -0.7005         0.5803    0.6683
s.e.    0.1402         0.1346    0.2901

sigma^2 = 0.2859:  log likelihood = -6.97
AIC=21.94   AICc=28.61   BIC=23.53

Training set error measures:
                   ME   RMSE      MAE       MPE     MAPE      MASE      ACF1
Training set -5528.76 100108 70922.88 -4.686453 30.29105 0.5123408 0.1401865
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag1_PDO
       -0.6351         0.4642    0.4817
s.e.    0.1370         0.1094    0.2669

sigma^2 = 0.2987:  log likelihood = -8.05
AIC=24.1   AICc=29.81   BIC=26.04

Training set error measures:
                   ME     RMSE      MAE        MPE     MAPE      MASE        ACF1
Training set 14842.41 80119.21 47134.71 -0.6707886 27.72089 0.2868815 -0.08239214
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag1_PDO
       -0.6944         0.3891    0.2791
s.e.    0.1432         0.1086    0.2596

sigma^2 = 0.3384:  log likelihood = -9.7
AIC=27.39   AICc=32.39   BIC=29.65

Training set error measures:
                   ME     RMSE      MAE       MPE     MAPE     MASE       ACF1
Training set 24051.02 97854.37 70453.41 0.9387005 35.38657 0.432233 -0.2089939
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag1_PDO
       -0.7107         0.3950    0.2983
s.e.    0.1330         0.1042    0.2473

sigma^2 = 0.3113:  log likelihood = -10.01
AIC=28.02   AICc=32.46   BIC=30.57

Training set error measures:
                   ME     RMSE      MAE        MPE     MAPE      MASE       ACF1
Training set 19912.29 94536.16 66207.64 -0.6065244 34.84564 0.4345582 -0.2307992
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag1_PDO
       -0.7781         0.3987    0.3633
s.e.    0.1384         0.1127    0.2647

sigma^2 = 0.3572:  log likelihood = -11.89
AIC=31.78   AICc=35.78   BIC=34.61

Training set error measures:
                   ME     RMSE      MAE      MPE     MAPE      MASE      ACF1
Training set 24574.21 108120.1 79245.38 2.487936 39.29337 0.5212155 -0.215193
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag1_PDO
       -0.7659         0.3940    0.3243
s.e.    0.1300         0.1087    0.2324

sigma^2 = 0.3323:  log likelihood = -12.23
AIC=32.46   AICc=36.09   BIC=35.55

Training set error measures:
                   ME     RMSE      MAE      MPE     MAPE      MASE       ACF1
Training set 26385.08 105818.9 75225.27 3.247573 37.37691 0.5089042 -0.2343519
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag1_PDO
       -0.7809         0.4050    0.3114
s.e.    0.1235         0.1039    0.2256

sigma^2 = 0.3132:  log likelihood = -12.6
AIC=33.21   AICc=36.54   BIC=36.54

Training set error measures:
                   ME     RMSE      MAE     MPE     MAPE      MASE       ACF1
Training set 31865.13 109992.6 81613.21 4.34429 37.12234 0.5143448 -0.2421801
Series: . 
Regression with ARIMA(0,0,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      intercept  lag1_log_SAR1
        15.8809         0.6363
s.e.     1.2594         0.1925

sigma^2 = 0.5826:  log likelihood = -8.04
AIC=22.08   AICc=28.08   BIC=22.32

Training set error measures:
                   ME     RMSE      MAE       MPE     MAPE     MASE       ACF1
Training set 21242.18 165773.4 120274.7 -24.54168 66.80343 1.047961 -0.3423622
Series: . 
Regression with ARIMA(0,0,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      intercept  lag1_log_SAR1
        15.8729         0.6354
s.e.     1.1604         0.1794

sigma^2 = 0.4994:  log likelihood = -8.51
AIC=23.03   AICc=27.83   BIC=23.62

Training set error measures:
                   ME     RMSE      MAE       MPE     MAPE     MASE       ACF1
Training set 19005.92 156012.2 107262.5 -21.75413 59.45741 1.089891 -0.3591053
Series: . 
Regression with ARIMA(0,0,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      intercept  lag1_log_SAR1
        16.3102         0.6942
s.e.     1.0788         0.1700

sigma^2 = 0.4869:  log likelihood = -9.48
AIC=24.95   AICc=28.95   BIC=25.86

Training set error measures:
                   ME     RMSE    MAE       MPE     MAPE     MASE       ACF1
Training set 23705.65 186927.1 127673 -23.52387 62.05895 1.297139 -0.3232275
Series: . 
Regression with ARIMA(0,0,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      intercept  lag1_log_SAR1
        16.5585         0.7229
s.e.     1.0754         0.1710

sigma^2 = 0.4921:  log likelihood = -10.6
AIC=27.21   AICc=30.64   BIC=28.4

Training set error measures:
                  ME     RMSE      MAE       MPE     MAPE     MASE       ACF1
Training set 25165.2 204748.5 139014.5 -25.49362 64.86593 1.124497 -0.1668198
Series: . 
Regression with ARIMA(1,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
          ar1  lag1_log_SAR1
      -0.6551         0.4049
s.e.   0.2006         0.1262

sigma^2 = 0.4541:  log likelihood = -10.44
AIC=26.89   AICc=30.31   BIC=28.08

Training set error measures:
                   ME     RMSE      MAE      MPE     MAPE      MASE       ACF1
Training set 72396.94 134771.9 101056.4 8.404581 39.93292 0.7300226 0.08189473
Series: . 
Regression with ARIMA(1,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
          ar1  lag1_log_SAR1
      -0.6647          0.420
s.e.   0.1901          0.096

sigma^2 = 0.4093:  log likelihood = -10.86
AIC=27.73   AICc=30.73   BIC=29.18

Training set error measures:
                   ME     RMSE      MAE     MPE     MAPE      MASE       ACF1
Training set 65376.58 130367.8 95450.82 7.00741 37.75224 0.5809535 0.07981914
Series: . 
Regression with ARIMA(1,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
          ar1  lag1_log_SAR1
      -0.6980         0.3598
s.e.   0.1937         0.0911

sigma^2 = 0.4455:  log likelihood = -12.44
AIC=30.88   AICc=33.54   BIC=32.57

Training set error measures:
                   ME     RMSE      MAE      MPE     MAPE      MASE       ACF1
Training set 69542.42 125779.1 94584.16 9.305403 41.46963 0.5802757 0.06005122
Series: . 
Regression with ARIMA(0,0,0)(0,1,0)[2] errors 
Box Cox transformation: lambda= 0 

Coefficients:
       drift  lag1_log_SAR1
      0.1706         0.4002
s.e.  0.0790         0.0717

sigma^2 = 0.3589:  log likelihood = -10.7
AIC=27.4   AICc=30.06   BIC=29.09

Training set error measures:
                    ME     RMSE     MAE       MPE     MAPE      MASE     ACF1
Training set -7674.584 142135.6 94775.7 -15.06531 46.45662 0.6220665 0.146008
Series: . 
Regression with ARIMA(0,0,0)(0,1,0)[2] errors 
Box Cox transformation: lambda= 0 

Coefficients:
       drift  lag1_log_SAR1
      0.1867         0.4002
s.e.  0.0753         0.0707

sigma^2 = 0.3447:  log likelihood = -11.33
AIC=28.66   AICc=31.06   BIC=30.58

Training set error measures:
                    ME     RMSE      MAE       MPE     MAPE      MASE      ACF1
Training set -7290.678 146519.4 98007.59 -15.11481 46.90963 0.6446189 0.1395689
Series: . 
Regression with ARIMA(0,0,0)(0,1,0)[2] errors 
Box Cox transformation: lambda= 0 

Coefficients:
       drift  lag1_log_SAR1
      0.2031         0.4051
s.e.  0.0716         0.0696

sigma^2 = 0.3329:  log likelihood = -11.96
AIC=29.92   AICc=32.1   BIC=32.05

Training set error measures:
                    ME     RMSE      MAE       MPE     MAPE      MASE      ACF1
Training set -11547.74 150205.8 97324.51 -15.17243 47.16615 0.6584071 0.1634555
Series: . 
Regression with ARIMA(0,0,0)(0,1,0)[2] errors 
Box Cox transformation: lambda= 0 

Coefficients:
       drift  lag1_log_SAR1
      0.1846         0.3882
s.e.  0.0677         0.0661

sigma^2 = 0.3241:  log likelihood = -12.62
AIC=31.24   AICc=33.24   BIC=33.56

Training set error measures:
                    ME     RMSE      MAE      MPE    MAPE      MASE       ACF1
Training set -22052.29 158497.9 108750.1 -14.5566 46.9133 0.6853676 0.06817126
Series: . 
Regression with ARIMA(0,0,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      intercept  pink_ind  lag1_NPGO
        11.9935   -1.2544     1.0341
s.e.     0.0960    0.1377     0.0761

sigma^2 = 0.05745:  log likelihood = 1.96
AIC=4.09   AICc=17.42   BIC=4.41

Training set error measures:
                   ME     RMSE      MAE       MPE     MAPE      MASE       ACF1
Training set 6855.985 55904.05 37552.36 -1.821041 15.68602 0.3271962 0.05367088
Series: . 
Regression with ARIMA(0,0,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      intercept  pink_ind  lag1_NPGO
        11.9963   -1.2736     1.0203
s.e.     0.0925    0.1294     0.0702

sigma^2 = 0.05015:  log likelihood = 2.52
AIC=2.96   AICc=12.96   BIC=3.75

Training set error measures:
                   ME     RMSE      MAE      MPE     MAPE     MASE        ACF1
Training set 6402.444 53578.12 35193.71 -1.67859 14.91594 0.357602 -0.05798536
Series: . 
Regression with ARIMA(0,0,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      intercept  pink_ind  lag1_NPGO
        12.1883   -1.4786     1.0368
s.e.     0.1488    0.2180     0.1252

sigma^2 = 0.1522:  log likelihood = -2.99
AIC=13.99   AICc=21.99   BIC=15.2

Training set error measures:
                   ME     RMSE      MAE       MPE     MAPE      MASE         ACF1
Training set 14440.23 117274.7 63104.08 -4.953999 22.03525 0.6411281 -0.005416942
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_NPGO
       -0.9915     0.5608
s.e.    0.2191     0.2379

sigma^2 = 0.4125:  log likelihood = -8.65
AIC=23.29   AICc=27.29   BIC=24.2

Training set error measures:
                   ME     RMSE    MAE      MPE     MAPE      MASE      ACF1
Training set 81468.78 161315.6 108218 6.954662 38.43314 0.8753826 0.1516856
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_NPGO
       -0.9631     0.6301
s.e.    0.2222     0.2356

sigma^2 = 0.42:  log likelihood = -9.73
AIC=25.47   AICc=28.89   BIC=26.66

Training set error measures:
                   ME     RMSE      MAE       MPE     MAPE      MASE       ACF1
Training set 51623.68 185816.7 131574.9 -0.537985 41.76673 0.9504855 -0.2232951
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_NPGO
       -0.9956     0.5276
s.e.    0.2425     0.2506

sigma^2 = 0.4941:  log likelihood = -11.7
AIC=29.41   AICc=32.41   BIC=30.86

Training set error measures:
                   ME     RMSE      MAE       MPE     MAPE      MASE        ACF1
Training set 24383.08 189573.6 132752.3 -17.09252 55.62664 0.8079857 -0.01464067
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_NPGO
       -1.0067     0.5266
s.e.    0.2288     0.2413

sigma^2 = 0.4511:  log likelihood = -12.19
AIC=30.37   AICc=33.04   BIC=32.07

Training set error measures:
                  ME     RMSE      MAE       MPE     MAPE      MASE        ACF1
Training set 23625.6 183530.1 124550.6 -14.95152 52.49028 0.7641201 -0.01746845
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_NPGO
       -0.9382     0.4530
s.e.    0.1903     0.1992

sigma^2 = 0.424:  log likelihood = -12.78
AIC=31.56   AICc=33.96   BIC=33.48

Training set error measures:
                   ME     RMSE      MAE       MPE     MAPE      MASE        ACF1
Training set 22491.73 177721.8 119651.1 -12.78704 52.21522 0.7853379 -0.03250186
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_NPGO
       -1.0063     0.4682
s.e.    0.1912     0.2062

sigma^2 = 0.4504:  log likelihood = -14.23
AIC=34.46   AICc=36.64   BIC=36.58

Training set error measures:
                   ME     RMSE      MAE       MPE     MAPE      MASE        ACF1
Training set 29634.26 182611.4 121540.8 -8.417347 51.97117 0.7994025 -0.03752842
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_NPGO
       -1.0083     0.4682
s.e.    0.1809     0.1996

sigma^2 = 0.4183:  log likelihood = -14.66
AIC=35.32   AICc=37.32   BIC=37.64

Training set error measures:
                   ME     RMSE      MAE       MPE     MAPE      MASE        ACF1
Training set 27464.08 177302.6 114556.7 -8.116634 49.07315 0.7749841 -0.04854465
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_NPGO
       -1.0137     0.4742
s.e.    0.1602     0.1771

sigma^2 = 0.3905:  log likelihood = -15.07
AIC=36.13   AICc=37.98   BIC=38.63

Training set error measures:
                   ME     RMSE    MAE       MPE     MAPE      MASE        ACF1
Training set 27438.42 172331.2 109437 -7.361133 46.47308 0.6896964 -0.04729607
Series: . 
Regression with ARIMA(0,0,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      intercept  pink_ind  lag1_log_SAR1  lag2_PDO
        15.1040   -0.5372         0.5145   -0.6603
s.e.     1.0325    0.3380         0.1421    0.3463

sigma^2 = 0.3944:  log likelihood = -4.86
AIC=19.72   AICc=49.72   BIC=20.11

Training set error measures:
                   ME     RMSE      MAE       MPE    MAPE      MASE       ACF1
Training set 9651.841 65992.59 55116.44 -10.31909 43.0291 0.4802331 -0.2183065
Series: . 
Regression with ARIMA(0,0,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      intercept  pink_ind  lag1_log_SAR1  lag2_PDO
        15.2974   -0.6454         0.5304    -0.504
s.e.     0.9643    0.2905         0.1360     0.254

sigma^2 = 0.3343:  log likelihood = -5.19
AIC=20.39   AICc=40.39   BIC=21.37

Training set error measures:
                   ME     RMSE     MAE       MPE     MAPE      MASE       ACF1
Training set 10148.38 65375.86 56277.8 -9.469089 40.70865 0.5718367 -0.1941175
Series: . 
Regression with ARIMA(0,0,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      intercept  pink_ind  lag1_log_SAR1  lag2_PDO
        15.3084   -0.6501         0.5317   -0.5079
s.e.     0.8965    0.2643         0.1273    0.2324

sigma^2 = 0.2787:  log likelihood = -5.25
AIC=20.49   AICc=35.49   BIC=22.01

Training set error measures:
                   ME     RMSE   MAE       MPE     MAPE      MASE       ACF1
Training set 9540.988 61694.33 51421 -8.526861 36.77015 0.5224297 -0.1994644
Series: . 
Regression with ARIMA(0,0,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      intercept  pink_ind  lag1_log_SAR1  lag2_PDO
        15.3925   -0.4618         0.5541   -0.5791
s.e.     1.0490    0.2937         0.1486    0.2696

sigma^2 = 0.3604:  log likelihood = -7.51
AIC=25.02   AICc=37.02   BIC=27.01

Training set error measures:
                   ME     RMSE      MAE       MPE    MAPE      MASE       ACF1
Training set 14964.74 115875.8 83276.62 -11.86156 44.3284 0.6736301 -0.1657804
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag2_PDO
       -0.6436         0.3791   -0.0472
s.e.    0.1677         0.1251    0.2589

sigma^2 = 0.4225:  log likelihood = -9.12
AIC=26.24   AICc=32.9   BIC=27.83

Training set error measures:
                   ME     RMSE     MAE       MPE     MAPE      MASE       ACF1
Training set 30766.87 129381.2 98769.2 -4.638975 40.42733 0.7135003 -0.1249158
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag2_PDO
       -0.6283         0.3607    -0.040
s.e.    0.1543         0.1062     0.248

sigma^2 = 0.3789:  log likelihood = -9.48
AIC=26.96   AICc=32.67   BIC=28.9

Training set error measures:
                   ME     RMSE      MAE       MPE     MAPE      MASE       ACF1
Training set 30824.95 122140.2 90812.88 -3.163218 38.01009 0.5527251 -0.1208698
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag2_PDO
       -0.6700         0.3380   -0.0625
s.e.    0.1473         0.1031    0.2460

sigma^2 = 0.3666:  log likelihood = -10.22
AIC=28.44   AICc=33.44   BIC=30.7

Training set error measures:
                   ME     RMSE     MAE       MPE     MAPE      MASE       ACF1
Training set 32041.21 119311.5 89788.2 -1.367951 39.51489 0.5508524 -0.1206652
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag2_PDO
       -0.6943         0.3440   -0.0149
s.e.    0.1392         0.1004    0.2291

sigma^2 = 0.3436:  log likelihood = -10.7
AIC=29.4   AICc=33.84   BIC=31.95

Training set error measures:
                   ME     RMSE      MAE       MPE     MAPE      MASE       ACF1
Training set 24533.37 120286.1 87624.98 -3.794704 39.56731 0.5751323 -0.1682337
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag2_PDO
       -0.7658         0.3345   -0.0672
s.e.    0.1462         0.1093    0.2478

sigma^2 = 0.4001:  log likelihood = -12.74
AIC=33.48   AICc=37.48   BIC=36.31

Training set error measures:
                   ME     RMSE      MAE       MPE    MAPE      MASE       ACF1
Training set 32422.67 130261.2 95792.81 -1.221467 43.0002 0.6300519 -0.1849444
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag2_PDO
       -0.7740         0.3328   -0.0740
s.e.    0.1372         0.1057    0.2385

sigma^2 = 0.3706:  log likelihood = -13.1
AIC=34.2   AICc=37.84   BIC=37.29

Training set error measures:
                  ME     RMSE      MAE       MPE     MAPE      MASE       ACF1
Training set 28842.1 126590.2 91204.76 -2.154896 41.43803 0.6170066 -0.2194397
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag2_PDO
       -0.7792         0.3373   -0.0896
s.e.    0.1301         0.0997    0.2154

sigma^2 = 0.3448:  log likelihood = -13.42
AIC=34.84   AICc=38.17   BIC=38.17

Training set error measures:
                   ME     RMSE      MAE       MPE     MAPE      MASE       ACF1
Training set 30337.32 123405.8 88959.87 -1.603965 39.79601 0.5606451 -0.2163525
Series: . 
Regression with ARIMA(0,0,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag4_log_adults
       -0.2138         0.1688           1.1656
s.e.    0.6950         0.2464           0.1434

sigma^2 = 1.575:  log likelihood = -11.29
AIC=30.58   AICc=43.91   BIC=30.9

Training set error measures:
                    ME     RMSE    MAE       MPE     MAPE     MASE      ACF1
Training set -58.20731 150632.5 116466 -36.34153 77.08295 1.014776 0.1419088
Series: . 
Regression with ARIMA(0,0,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag4_log_adults
       -0.3585         0.1248           1.1400
s.e.    0.6378         0.2295           0.1335

sigma^2 = 1.383:  log likelihood = -12.41
AIC=32.81   AICc=42.81   BIC=33.6

Training set error measures:
                    ME     RMSE      MAE       MPE     MAPE   MASE      ACF1
Training set -278.7649 139939.8 112085.8 -32.69167 73.61077 1.1389 0.3509215
Series: . 
Regression with ARIMA(0,0,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      intercept  pink_ind  lag1_log_SAR1  lag4_log_adults
        17.1335   -0.7773         0.6820          -0.0448
s.e.     3.4270    0.3314         0.1638           0.2367

sigma^2 = 0.4103:  log likelihood = -7.18
AIC=24.36   AICc=39.36   BIC=25.88

Training set error measures:
                   ME     RMSE      MAE      MPE     MAPE      MASE        ACF1
Training set 13518.32 107180.3 78429.34 -13.2544 45.62017 0.7968306 -0.09906304
Series: . 
Regression with ARIMA(0,0,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      intercept  pink_ind  lag1_log_SAR1  lag4_log_adults
        15.2895   -0.5259         0.6677            0.106
s.e.     3.7772    0.3520         0.1867            0.257

sigma^2 = 0.5037:  log likelihood = -9.35
AIC=28.7   AICc=40.7   BIC=30.69

Training set error measures:
                   ME     RMSE      MAE       MPE     MAPE      MASE        ACF1
Training set 17684.47 147480.6 99141.74 -16.80247 47.82345 0.8019641 -0.08501487
Series: . 
Regression with ARIMA(0,0,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      intercept  pink_ind  lag1_log_SAR1  lag4_log_adults
        12.0783   -0.5637         0.5293           0.3174
s.e.     2.9237    0.3605         0.1572           0.2038

sigma^2 = 0.5077:  log likelihood = -10.53
AIC=31.05   AICc=41.05   BIC=33.48

Training set error measures:
                   ME     RMSE      MAE       MPE     MAPE      MASE       ACF1
Training set 16376.28 132614.1 106235.3 -16.41429 51.32813 0.7674344 0.01332912
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag4_log_adults
       -0.6121         0.3530           0.0307
s.e.    0.2011         0.1301           0.2448

sigma^2 = 0.3792:  log likelihood = -9.48
AIC=26.97   AICc=32.68   BIC=28.91

Training set error measures:
                   ME     RMSE      MAE       MPE     MAPE      MASE       ACF1
Training set 26271.01 127842.6 94486.87 -3.216393 37.46435 0.5750865 -0.1386195
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag4_log_adults
       -0.6032         0.3101           0.1120
s.e.    0.1985         0.1183           0.2222

sigma^2 = 0.3614:  log likelihood = -10.12
AIC=28.25   AICc=33.25   BIC=30.51

Training set error measures:
                   ME     RMSE      MAE       MPE     MAPE      MASE       ACF1
Training set 19030.69 139462.8 100706.1 -2.147349 38.23752 0.6178336 -0.1343157
Series: . 
Regression with ARIMA(0,0,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      intercept  pink_ind  lag1_log_SAR1  lag4_log_adults
         8.7608   -0.4578         0.3146           0.4868
s.e.     1.8872    0.3130         0.0779           0.1664

sigma^2 = 0.4576:  log likelihood = -13.09
AIC=36.19   AICc=42.86   BIC=39.73

Training set error measures:
                   ME     RMSE      MAE       MPE     MAPE      MASE          ACF1
Training set 20156.48 105217.1 74112.98 -17.22548 48.72597 0.4864454 -0.0003307733
Series: . 
Regression with ARIMA(0,0,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      intercept  pink_ind  lag1_log_SAR1  lag4_log_adults
         8.5581   -0.6243         0.2441           0.4740
s.e.     2.0595    0.3301         0.0761           0.1817

sigma^2 = 0.5343:  log likelihood = -15.39
AIC=40.78   AICc=46.78   BIC=44.64

Training set error measures:
                   ME     RMSE      MAE       MPE     MAPE      MASE        ACF1
Training set 24857.38 117994.4 82344.89 -21.16813 52.88874 0.5416017 -0.07023061
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag4_log_adults
       -0.7966         0.3448          -0.0399
s.e.    0.1782         0.1179           0.2008

sigma^2 = 0.3719:  log likelihood = -13.13
AIC=34.26   AICc=37.89   BIC=37.35

Training set error measures:
                   ME     RMSE      MAE       MPE     MAPE      MASE      ACF1
Training set 28575.76 127198.3 92264.71 -1.759894 41.89407 0.6241772 -0.264041
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag4_log_adults
       -0.8082         0.3545          -0.0435
s.e.    0.1692         0.1103           0.1950

sigma^2 = 0.3473:  log likelihood = -13.48
AIC=34.96   AICc=38.3   BIC=38.3

Training set error measures:
                   ME     RMSE      MAE        MPE     MAPE      MASE       ACF1
Training set 31781.64 126218.5 92965.98 -0.7038464 40.69059 0.5858925 -0.2784653
Series: . 
Regression with ARIMA(0,0,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      intercept  pink_ind  lag1_log_SAR1  lag2_NPGO
        15.1164   -0.6318         0.4939     0.3907
s.e.     1.0058    0.3181         0.1439     0.1962

sigma^2 = 0.3836:  log likelihood = -4.75
AIC=19.49   AICc=49.49   BIC=19.89

Training set error measures:
                   ME     RMSE      MAE       MPE     MAPE      MASE       ACF1
Training set 9406.182 71192.19 57597.17 -10.02184 37.98034 0.5018479 -0.3653301
Series: . 
Regression with ARIMA(0,0,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      intercept  pink_ind  lag1_log_SAR1  lag2_NPGO
        15.1303   -0.5873         0.4974     0.4040
s.e.     0.9575    0.2847         0.1368     0.1843

sigma^2 = 0.3133:  log likelihood = -4.9
AIC=19.8   AICc=39.8   BIC=20.79

Training set error measures:
                   ME     RMSE      MAE       MPE     MAPE      MASE       ACF1
Training set 7671.213 70429.12 56540.41 -9.228853 36.58432 0.5745051 -0.4202522
Series: . 
Regression with ARIMA(0,0,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      intercept  pink_ind  lag1_log_SAR1  lag2_NPGO
        15.2011   -0.6109         0.5053     0.4120
s.e.     0.8824    0.2600         0.1278     0.1737

sigma^2 = 0.2635:  log likelihood = -4.97
AIC=19.93   AICc=34.93   BIC=21.45

Training set error measures:
                   ME     RMSE      MAE       MPE     MAPE     MASE       ACF1
Training set 8562.848 70407.13 58537.66 -8.470093 34.59766 0.594734 -0.4036727
Series: . 
Regression with ARIMA(0,0,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      intercept  pink_ind  lag1_log_SAR1  lag2_NPGO
        15.6799   -0.4135         0.5766     0.3600
s.e.     1.1179    0.3235         0.1618     0.2228

sigma^2 = 0.4134:  log likelihood = -8.26
AIC=26.53   AICc=38.53   BIC=28.52

Training set error measures:
                   ME     RMSE      MAE       MPE     MAPE      MASE       ACF1
Training set 15715.96 138305.2 94111.07 -13.58249 43.71144 0.7612707 -0.1378387
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag2_NPGO
       -0.5451         0.3045     0.2244
s.e.    0.1918         0.1447     0.2380

sigma^2 = 0.3921:  log likelihood = -8.71
AIC=25.41   AICc=32.08   BIC=27.01

Training set error measures:
                   ME     RMSE      MAE         MPE     MAPE      MASE       ACF1
Training set 33157.28 125454.2 96001.03 -0.06639832 38.92715 0.6935033 -0.1997998
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag2_NPGO
       -0.5312         0.2881     0.2269
s.e.    0.1777         0.1263     0.2284

sigma^2 = 0.3509:  log likelihood = -9.02
AIC=26.03   AICc=31.75   BIC=27.97

Training set error measures:
                   ME     RMSE      MAE      MPE     MAPE      MASE       ACF1
Training set 32840.41 119141.6 88210.05 0.923707 36.57487 0.5368832 -0.1879436
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag2_NPGO
       -0.5296         0.2627     0.2744
s.e.    0.1724         0.1117     0.2005

sigma^2 = 0.3221:  log likelihood = -9.38
AIC=26.75   AICc=31.75   BIC=29.01

Training set error measures:
                  ME     RMSE      MAE      MPE     MAPE      MASE       ACF1
Training set 32467.5 114459.1 83794.58 2.480709 35.44268 0.5140814 -0.1937456
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag2_NPGO
       -0.5503          0.266     0.2767
s.e.    0.1649          0.109     0.1958

sigma^2 = 0.3008:  log likelihood = -9.77
AIC=27.53   AICc=31.98   BIC=30.09

Training set error measures:
                   ME     RMSE      MAE           MPE     MAPE     MASE       ACF1
Training set 27351.36 111118.7 80101.75 -0.0006188346 35.65753 0.525753 -0.1944934
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag2_NPGO
       -0.7538         0.3287     0.0257
s.e.    0.1679         0.1237     0.1968

sigma^2 = 0.4016:  log likelihood = -12.77
AIC=33.54   AICc=37.54   BIC=36.37

Training set error measures:
                   ME     RMSE      MAE         MPE     MAPE      MASE       ACF1
Training set 30713.11 134832.4 98044.72 -0.08276243 42.28519 0.6448632 -0.2171674
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag2_NPGO
       -0.7631         0.3268      0.026
s.e.    0.1591         0.1199      0.191

sigma^2 = 0.3724:  log likelihood = -13.14
AIC=34.28   AICc=37.91   BIC=37.37

Training set error measures:
                   ME     RMSE      MAE       MPE    MAPE      MASE       ACF1
Training set 26694.13 131801.3 93920.89 -1.173038 40.9237 0.6353814 -0.2545317
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag2_NPGO
       -0.7768         0.3386     0.0182
s.e.    0.1494         0.1113     0.1845

sigma^2 = 0.3481:  log likelihood = -13.5
AIC=35   AICc=38.34   BIC=38.34

Training set error measures:
                   ME   RMSE      MAE      MPE     MAPE      MASE       ACF1
Training set 29941.72 131978 95408.98 -0.21586 39.92634 0.6012889 -0.2633054
Series: . 
Regression with ARIMA(0,0,0)(1,0,0)[2] errors 
Box Cox transformation: lambda= 0 

Coefficients:
         sar1  intercept  pink_ind  lag1_log_SAR1  lag2_PDO  lag1_NPGO  lag1_PDO
      -0.3485     8.8885   -1.3286        -0.4021    0.3470     1.1554   -0.9765
s.e.   0.5176     2.0972    0.1563         0.2825    0.2963     0.1846    0.5872

sigma^2 = 0.1802:  log likelihood = 3.69
AIC=8.62   AICc=-135.38   BIC=9.25

Training set error measures:
                   ME    RMSE      MAE        MPE    MAPE      MASE       ACF1
Training set 3314.691 35052.9 26837.76 -0.3739742 12.5008 0.2338392 -0.3432453
Series: . 
Regression with ARIMA(1,0,0)(1,0,0)[2] errors 
Box Cox transformation: lambda= 0 

Coefficients:
         ar1     sar1  intercept  pink_ind  lag1_log_SAR1  lag2_PDO  lag1_NPGO  lag1_PDO
      -0.904  -0.8834     8.5434   -1.4921        -0.4581    0.5942     1.3110   -1.0983
s.e.   0.140   0.1502     0.8641    0.2564         0.1218    0.2740     0.1387    0.3308

sigma^2 = 0.1005:  log likelihood = 5.64
AIC=6.73   AICc=-173.27   BIC=8.5

Training set error measures:
                   ME     RMSE      MAE      MPE     MAPE      MASE       ACF1
Training set 5841.781 27034.26 19395.91 2.068481 9.497489 0.1970811 -0.6077804
Series: . 
Regression with ARIMA(0,0,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag2_PDO  lag1_NPGO  lag1_PDO
       -1.9753        -1.6057    0.5521     1.4322   -3.2676
s.e.    0.8292         0.1154    0.7263     0.7116    1.0439

sigma^2 = 2.072:  log likelihood = -14.37
AIC=40.73   AICc=68.73   BIC=42.55

Training set error measures:
                    ME     RMSE      MAE       MPE     MAPE     MASE       ACF1
Training set -63212.48 303163.5 206010.3 -35.05266 77.72057 2.093035 0.05199457
Series: . 
Regression with ARIMA(0,0,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      intercept  pink_ind  lag1_log_SAR1  lag2_PDO  lag1_NPGO  lag1_PDO
        16.2517   -0.9774         0.5798   -0.6129     0.6600    1.1367
s.e.     0.7380    0.2108         0.1056    0.1599     0.1956    0.2398

sigma^2 = 0.1657:  log likelihood = -1.39
AIC=16.77   AICc=54.1   BIC=19.56

Training set error measures:
                   ME     RMSE     MAE       MPE     MAPE      MASE      ACF1
Training set 8862.881 81340.01 53741.4 -3.693213 22.29141 0.4347177 0.2072301
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag2_PDO  lag1_NPGO  lag1_PDO
       -1.0871         0.4876   -0.1012     0.6542    1.0337
s.e.    0.1013         0.0721    0.1126     0.1224    0.1640

sigma^2 = 0.1008:  log likelihood = 0.35
AIC=11.31   AICc=32.31   BIC=13.69

Training set error measures:
                   ME     RMSE      MAE      MPE     MAPE     MASE        ACF1
Training set 28481.39 63417.76 40959.92 4.484188 15.91442 0.295891 -0.05780253
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag2_PDO  lag1_NPGO  lag1_PDO
       -1.0879         0.4439   -0.0804     0.6925    0.9884
s.e.    0.1006         0.0539    0.1096     0.1143    0.1554

sigma^2 = 0.09306:  log likelihood = 0.46
AIC=11.09   AICc=27.89   BIC=14

Training set error measures:
                   ME     RMSE      MAE      MPE     MAPE     MASE       ACF1
Training set 32364.15 67317.19 44955.68 5.742479 16.29521 0.273619 -0.2021448
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag2_PDO  lag1_NPGO  lag1_PDO
       -1.0953         0.3505   -0.0953     0.5874    0.6570
s.e.    0.1625         0.0809    0.1769     0.1810    0.2233

sigma^2 = 0.2299:  log likelihood = -5.73
AIC=23.47   AICc=37.47   BIC=26.86

Training set error measures:
                   ME     RMSE      MAE      MPE     MAPE      MASE       ACF1
Training set 41887.37 86295.06 61683.54 7.657282 24.25683 0.3784297 -0.1273726
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag2_PDO  lag1_NPGO  lag1_PDO
       -0.9431         0.3489   -0.1571     0.4143    0.5225
s.e.    0.1334         0.0846    0.1800     0.1467    0.2142

sigma^2 = 0.2404:  log likelihood = -6.79
AIC=25.59   AICc=37.59   BIC=29.42

Training set error measures:
                   ME     RMSE      MAE      MPE     MAPE    MASE      ACF1
Training set 42758.03 83550.01 60898.32 8.234629 27.86921 0.39971 -0.123847
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag2_PDO  lag1_NPGO  lag1_PDO
       -1.0184         0.3481   -0.2132     0.4528    0.6034
s.e.    0.1369         0.0911    0.1911     0.1565    0.2259

sigma^2 = 0.2691:  log likelihood = -8.4
AIC=28.79   AICc=39.29   BIC=33.04

Training set error measures:
                   ME     RMSE      MAE      MPE     MAPE      MASE       ACF1
Training set 48453.11 86204.55 64922.98 10.83479 30.38286 0.4270137 -0.1124438
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag2_PDO  lag1_NPGO  lag1_PDO
       -0.9718         0.3398   -0.1681     0.4218    0.4860
s.e.    0.1307         0.0913    0.1876     0.1547    0.1994

sigma^2 = 0.2638:  log likelihood = -9.05
AIC=30.09   AICc=39.42   BIC=34.73

Training set error measures:
                   ME     RMSE      MAE      MPE     MAPE      MASE        ACF1
Training set 50882.52 92868.05 67357.66 11.89635 29.97988 0.4556793 -0.09501997
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag2_PDO  lag1_NPGO  lag1_PDO
       -0.9441          0.334   -0.1275     0.3916    0.4819
s.e.    0.1196          0.089    0.1710     0.1431    0.1954

sigma^2 = 0.2471:  log likelihood = -9.28
AIC=30.55   AICc=38.95   BIC=35.55

Training set error measures:
                   ME     RMSE      MAE     MPE     MAPE      MASE       ACF1
Training set 38337.04 97044.61 73551.52 9.55554 29.88653 0.4635382 -0.1019952
Series: . 
Regression with ARIMA(1,0,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
          ar1  intercept  pink_ind  lag1_log_SAR1  lag2_NPGO  lag1_NPGO  lag1_PDO
      -0.9209    10.7918   -1.1045        -0.1353     0.1571     0.9386   -0.3170
s.e.   0.1151     0.3105    0.2730         0.0351     0.0211     0.0473    0.1034

sigma^2 = 0.02643:  log likelihood = 10.56
AIC=-5.11   AICc=-149.11   BIC=-4.48

Training set error measures:
                   ME     RMSE      MAE      MPE    MAPE       MASE     ACF1
Training set 3613.013 15352.17 10696.26 0.505158 5.37932 0.09319721 -0.24009
Series: . 
Regression with ARIMA(1,0,0)(1,0,0)[2] errors 
Box Cox transformation: lambda= 0 

Coefficients:
          ar1    sar1  intercept  pink_ind  lag1_log_SAR1  lag2_NPGO  lag1_NPGO  lag1_PDO
      -0.9440  0.6519    10.8933   -1.3234        -0.1411     0.1484     0.9834   -0.2443
s.e.   0.0792  0.3029     0.3695    0.6712         0.0209     0.0122     0.0248    0.0506

sigma^2 = 0.01894:  log likelihood = 12.43
AIC=-6.87   AICc=-186.87   BIC=-5.09

Training set error measures:
                   ME     RMSE      MAE       MPE     MAPE       MASE       ACF1
Training set 3493.392 17038.43 8356.945 0.1919653 3.515938 0.08491462 -0.3202011
Series: . 
Regression with ARIMA(0,0,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      intercept  pink_ind  lag1_log_SAR1  lag2_NPGO  lag1_NPGO  lag1_PDO
        13.5100   -1.1464         0.2134     0.2956     0.7452    0.2319
s.e.     0.9325    0.1793         0.1255     0.0998     0.1562    0.2874

sigma^2 = 0.1172:  log likelihood = 1.11
AIC=11.78   AICc=67.78   BIC=13.9

Training set error measures:
                  ME     RMSE      MAE       MPE     MAPE      MASE       ACF1
Training set 11653.1 72649.71 39497.74 -2.293218 16.87068 0.4012912 0.06647362
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag2_NPGO  lag1_NPGO  lag1_PDO
       -0.9755         0.4321     0.2058     0.6133    1.0363
s.e.    0.1098         0.0719     0.1010     0.1127    0.1496

sigma^2 = 0.08976:  log likelihood = 1.33
AIC=9.34   AICc=37.34   BIC=11.15

Training set error measures:
                   ME     RMSE      MAE      MPE     MAPE      MASE       ACF1
Training set 28841.97 62759.08 39467.68 8.199871 13.86348 0.3192567 -0.1582691
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag2_NPGO  lag1_NPGO  lag1_PDO
       -0.9733         0.4340     0.2078     0.6136    1.0388
s.e.    0.1017         0.0651     0.0938     0.1074    0.1400

sigma^2 = 0.07485:  log likelihood = 1.99
AIC=8.03   AICc=29.03   BIC=10.42

Training set error measures:
                   ME     RMSE     MAE      MPE     MAPE      MASE       ACF1
Training set 25814.35 60177.26 36580.9 7.369437 12.75736 0.2642573 -0.1792715
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag2_NPGO  lag1_NPGO  lag1_PDO
       -0.9811         0.3913     0.1946     0.6559    0.9933
s.e.    0.1027         0.0539     0.0943     0.1021    0.1359

sigma^2 = 0.07176:  log likelihood = 2.02
AIC=7.97   AICc=24.77   BIC=10.88

Training set error measures:
                   ME     RMSE     MAE      MPE     MAPE      MASE      ACF1
Training set 30766.82 64145.23 40071.5 8.339547 13.92051 0.2438918 -0.306813
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag2_NPGO  lag1_NPGO  lag1_PDO
       -0.9064         0.2823     0.3394     0.5491    0.7581
s.e.    0.1538         0.0721     0.1336     0.1504    0.1886

sigma^2 = 0.1571:  log likelihood = -3.26
AIC=18.51   AICc=32.51   BIC=21.9

Training set error measures:
                   ME     RMSE      MAE      MPE     MAPE      MASE       ACF1
Training set 37605.13 76036.65 51838.98 11.44129 19.56333 0.3180332 -0.2742785
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag2_NPGO  lag1_NPGO  lag1_PDO
       -0.7339         0.2806     0.3402     0.3539    0.5931
s.e.    0.1417         0.0797     0.1477     0.1279    0.1882

sigma^2 = 0.1839:  log likelihood = -4.92
AIC=21.83   AICc=33.83   BIC=25.67

Training set error measures:
                   ME     RMSE      MAE      MPE     MAPE      MASE       ACF1
Training set 37411.72 74836.34 56663.02 12.20265 24.85157 0.3719114 -0.3711134
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag2_NPGO  lag1_NPGO  lag1_PDO
       -0.9555         0.3270     0.1045     0.4226    0.5943
s.e.    0.1558         0.0991     0.1613     0.1593    0.2380

sigma^2 = 0.2835:  log likelihood = -8.79
AIC=29.58   AICc=40.08   BIC=33.82

Training set error measures:
                  ME     RMSE      MAE      MPE     MAPE      MASE       ACF1
Training set 44532.8 93406.55 68380.07 13.33251 29.33285 0.4497518 -0.2191264
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag2_NPGO  lag1_NPGO  lag1_PDO
       -0.9271         0.3242     0.0817     0.4009     0.490
s.e.    0.1521         0.0986     0.1587     0.1569     0.209

sigma^2 = 0.2726:  log likelihood = -9.31
AIC=30.61   AICc=39.94   BIC=35.25

Training set error measures:
                   ME     RMSE      MAE      MPE     MAPE      MASE       ACF1
Training set 47218.08 97179.33 69412.84 13.57033 29.08092 0.4695827 -0.1979881
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag2_NPGO  lag1_NPGO  lag1_PDO
       -0.9165         0.3212     0.0868     0.3919    0.4926
s.e.    0.1351         0.0943     0.1515     0.1440    0.2024

sigma^2 = 0.2503:  log likelihood = -9.39
AIC=30.78   AICc=39.18   BIC=35.78

Training set error measures:
                   ME     RMSE      MAE      MPE     MAPE      MASE       ACF1
Training set 42015.08 94775.35 68226.77 12.37085 27.93514 0.4299804 -0.1870869
Series: . 
Regression with ARIMA(1,0,0)(1,0,0)[2] errors 
Box Cox transformation: lambda= 0 

Coefficients:
         ar1    sar1  intercept  pink_ind  lag1_log_SAR1  lag2_PDO  lag1_NPGO
      0.3912  0.1878    12.2697   -1.1937         0.0392   -0.0725     0.9596
s.e.  0.4285  0.4566     0.5253    0.1447         0.0770    0.1646     0.1297

sigma^2 = 0.2526:  log likelihood = 2.32
AIC=11.35   AICc=-132.65   BIC=11.99

Training set error measures:
                   ME     RMSE      MAE       MPE    MAPE      MASE        ACF1
Training set 9161.842 45913.82 33714.21 -3.474982 16.8767 0.2937542 -0.07410826
Series: . 
Regression with ARIMA(0,0,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      intercept  pink_ind  lag1_log_SAR1  lag2_PDO  lag1_NPGO
        12.0681   -1.2739         0.0096    0.0164     1.0172
s.e.     0.6481    0.1573         0.0995    0.1347     0.1585

sigma^2 = 0.07494:  log likelihood = 2.54
AIC=6.92   AICc=48.92   BIC=8.11

Training set error measures:
                   ME    RMSE     MAE       MPE     MAPE      MASE        ACF1
Training set 6378.012 53539.5 35738.4 -1.684525 15.07052 0.3631366 -0.02271966
Series: . 
Regression with ARIMA(0,0,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      intercept  pink_ind  lag1_log_SAR1  lag2_PDO  lag1_NPGO
        13.3388   -1.1817         0.2035   -0.2353     0.6955
s.e.     0.8202    0.2325         0.1263    0.1736     0.1977

sigma^2 = 0.1495:  log likelihood = -1.22
AIC=14.44   AICc=42.44   BIC=16.26

Training set error measures:
                  ME     RMSE      MAE       MPE     MAPE      MASE        ACF1
Training set 11739.6 79914.83 50745.17 -3.576857 22.88195 0.5155635 -0.09041401
Series: . 
Regression with ARIMA(0,0,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      intercept  pink_ind  lag1_log_SAR1  lag2_PDO  lag1_NPGO
        15.3413   -0.4710         0.5458   -0.5737     0.0189
s.e.     1.2430    0.3171         0.1838    0.2786     0.2465

sigma^2 = 0.4202:  log likelihood = -7.51
AIC=27.01   AICc=48.01   BIC=29.4

Training set error measures:
                   ME     RMSE      MAE       MPE     MAPE     MASE       ACF1
Training set 15184.34 116731.8 82590.01 -11.80543 43.88823 0.668076 -0.1527225
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag2_PDO  lag1_NPGO
       -0.8441         0.2667    0.0146     0.3735
s.e.    0.2012         0.1354    0.2387     0.2448

sigma^2 = 0.3985:  log likelihood = -8.06
AIC=26.12   AICc=38.12   BIC=28.11

Training set error measures:
                   ME     RMSE     MAE      MPE     MAPE      MASE        ACF1
Training set 48970.05 145219.5 98354.3 1.925781 32.22265 0.7105031 -0.05495904
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag2_PDO  lag1_NPGO
       -0.8308         0.3062   -0.0037     0.3149
s.e.    0.1927         0.1033    0.2278     0.2043

sigma^2 = 0.3558:  log likelihood = -8.39
AIC=26.79   AICc=36.79   BIC=29.21

Training set error measures:
                   ME     RMSE      MAE        MPE    MAPE      MASE       ACF1
Training set 40387.24 136606.2 92854.09 -0.4029783 32.2943 0.5651487 0.01438316
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag2_PDO  lag1_NPGO
       -0.8757         0.2835   -0.0232     0.3252
s.e.    0.1862         0.1002    0.2262     0.2033

sigma^2 = 0.3404:  log likelihood = -9.05
AIC=28.1   AICc=36.67   BIC=30.93

Training set error measures:
                   ME     RMSE      MAE      MPE     MAPE      MASE        ACF1
Training set 40838.16 132963.6 91228.15 1.419078 33.55595 0.5596865 0.008209864
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag2_PDO  lag1_NPGO
       -0.8466         0.2868   -0.0439     0.2929
s.e.    0.1521         0.0963    0.2076     0.1648

sigma^2 = 0.3083:  log likelihood = -9.27
AIC=28.55   AICc=36.05   BIC=31.74

Training set error measures:
                   ME     RMSE      MAE      MPE     MAPE      MASE        ACF1
Training set 39184.58 126242.2 84803.93 1.912062 32.39558 0.5566161 0.005138645
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag2_PDO  lag1_NPGO
       -0.9266         0.2729   -0.0951     0.3190
s.e.    0.1610         0.1053    0.2259     0.1801

sigma^2 = 0.361:  log likelihood = -11.32
AIC=32.63   AICc=39.3   BIC=36.17

Training set error measures:
                   ME     RMSE      MAE      MPE     MAPE      MASE       ACF1
Training set 47015.34 132392.7 91467.07 4.846297 34.98242 0.6016004 0.01144214
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag2_PDO  lag1_NPGO
       -0.9333         0.2714   -0.1003     0.3201
s.e.    0.1519         0.1017    0.2172     0.1744

sigma^2 = 0.3317:  log likelihood = -11.57
AIC=33.14   AICc=39.14   BIC=37.01

Training set error measures:
                   ME     RMSE      MAE      MPE     MAPE      MASE        ACF1
Training set 43067.72 128358.5 86757.32 3.880549 33.54867 0.5869194 -0.02445242
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag2_PDO  lag1_NPGO
       -0.9099         0.2670   -0.0659     0.2949
s.e.    0.1384         0.0988    0.1972     0.1603

sigma^2 = 0.3097:  log likelihood = -11.88
AIC=33.76   AICc=39.21   BIC=37.92

Training set error measures:
                   ME     RMSE      MAE      MPE     MAPE      MASE        ACF1
Training set 32572.67 128862.7 90015.43 2.204418 33.09809 0.5672975 -0.02121208
Series: . 
Regression with ARIMA(1,0,0)(1,0,0)[2] errors 
Box Cox transformation: lambda= 0 

Coefficients:
         ar1     sar1  intercept  pink_ind  lag1_log_SAR1  lag1_NPGO  lag4_log_adults
      0.2910  -0.2080    10.0804   -1.1905        -0.0321     1.0493           0.1508
s.e.  0.5256   0.7599     3.2893    0.1158         0.1303     0.2353           0.2295

sigma^2 = 0.2392:  log likelihood = 2.62
AIC=10.76   AICc=-133.24   BIC=11.4

Training set error measures:
                   ME     RMSE      MAE       MPE     MAPE      MASE      ACF1
Training set 6491.588 53997.87 34770.28 -1.749975 12.85913 0.3029557 0.0225349
Series: . 
Regression with ARIMA(0,0,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag1_NPGO  lag4_log_adults
       -1.3474        -0.5960     1.5081           0.7105
s.e.    0.3750         0.1819     0.2937           0.1074

sigma^2 = 0.4223:  log likelihood = -6.25
AIC=22.49   AICc=42.49   BIC=23.48

Training set error measures:
                   ME     RMSE      MAE       MPE     MAPE      MASE        ACF1
Training set -9466.54 124329.7 93077.96 -11.11775 44.99104 0.9457618 -0.09955594
Series: . 
Regression with ARIMA(0,0,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag1_NPGO  lag4_log_adults
       -1.3617        -0.5054     1.3992           0.7712
s.e.    0.3727         0.1539     0.2687           0.0856

sigma^2 = 0.3869:  log likelihood = -6.89
AIC=23.78   AICc=38.78   BIC=25.29

Training set error measures:
                    ME     RMSE      MAE       MPE     MAPE    MASE       ACF1
Training set -4393.327 136580.8 100801.6 -11.63788 44.84258 1.02413 -0.2164616
Series: . 
Regression with ARIMA(0,0,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag1_NPGO  lag4_log_adults
       -0.5740        -0.1700     0.7704           0.9597
s.e.    0.4664         0.1899     0.3194           0.1052

sigma^2 = 0.8201:  log likelihood = -12.03
AIC=34.06   AICc=46.06   BIC=36.05

Training set error measures:
                    ME     RMSE      MAE       MPE     MAPE    MASE      ACF1
Training set -209.6096 183738.4 136136.9 -22.96127 59.76629 1.10122 0.1175844
Series: . 
Regression with ARIMA(0,0,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag1_NPGO  lag4_log_adults
       -0.5693        -0.1731     0.7850           0.9568
s.e.    0.4442         0.1793     0.2685           0.0965

sigma^2 = 0.7182:  log likelihood = -12.61
AIC=35.22   AICc=45.22   BIC=37.64

Training set error measures:
                    ME     RMSE    MAE       MPE     MAPE      MASE       ACF1
Training set -418.6973 176235.3 125012 -21.00367 54.59898 0.9030757 0.09691145
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag1_NPGO  lag4_log_adults
       -0.8519         0.3156     0.3207          -0.0332
s.e.    0.2398         0.1211     0.2064           0.2271

sigma^2 = 0.3552:  log likelihood = -8.38
AIC=26.77   AICc=36.77   BIC=29.19

Training set error measures:
                   ME     RMSE      MAE        MPE    MAPE     MASE       ACF1
Training set 43082.61 134566.3 90093.89 -0.1204836 32.0815 0.548349 0.02024548
Series: . 
Regression with ARIMA(0,0,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      intercept  pink_ind  lag1_log_SAR1  lag1_NPGO  lag4_log_adults
         7.2929   -0.6432         0.1833     0.4001           0.5371
s.e.     2.0693    0.3262         0.1060     0.2180           0.1631

sigma^2 = 0.4502:  log likelihood = -11.19
AIC=34.37   AICc=46.37   BIC=38.21

Training set error measures:
                   ME   RMSE     MAE       MPE     MAPE      MASE      ACF1
Training set 15862.63 128275 94614.1 -13.29208 44.96024 0.5804594 0.1704881
Series: . 
Regression with ARIMA(0,0,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      intercept  pink_ind  lag1_log_SAR1  lag1_NPGO  lag4_log_adults
         7.1231   -0.6105         0.1780     0.3868           0.5478
s.e.     1.9078    0.2930         0.1009     0.2056           0.1531

sigma^2 = 0.4073:  log likelihood = -11.51
AIC=35.01   AICc=45.51   BIC=39.26

Training set error measures:
                   ME     RMSE      MAE       MPE     MAPE      MASE      ACF1
Training set 15569.49 122297.9 86055.16 -12.54898 42.14659 0.5648286 0.1644781
Series: . 
Regression with ARIMA(0,0,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      intercept  pink_ind  lag1_log_SAR1  lag1_NPGO  lag4_log_adults
         7.1695   -0.7666         0.1235     0.3237           0.5240
s.e.     2.1906    0.3285         0.1130     0.2342           0.1755

sigma^2 = 0.5208:  log likelihood = -14.49
AIC=40.97   AICc=50.3   BIC=45.61

Training set error measures:
                   ME     RMSE      MAE       MPE     MAPE      MASE       ACF1
Training set 21578.32 128494.8 90890.86 -16.55365 47.56531 0.5978106 0.06101985
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag1_NPGO  lag4_log_adults
       -0.9586         0.2870     0.3160          -0.0484
s.e.    0.1855         0.1121     0.1749           0.1831

sigma^2 = 0.3346:  log likelihood = -11.64
AIC=33.29   AICc=39.29   BIC=37.15

Training set error measures:
                   ME     RMSE      MAE      MPE     MAPE      MASE        ACF1
Training set 42653.69 129044.6 87454.57 4.259913 34.29758 0.5916363 -0.09315562
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag1_NPGO  lag4_log_adults
       -0.9399         0.2821     0.2986          -0.0449
s.e.    0.1697         0.1078     0.1602           0.1777

sigma^2 = 0.3106:  log likelihood = -11.9
AIC=33.8   AICc=39.26   BIC=37.97

Training set error measures:
                   ME     RMSE      MAE      MPE     MAPE      MASE        ACF1
Training set 34885.43 127144.6 87827.67 2.861674 33.33697 0.5535097 -0.06138387
Series: . 
Regression with ARIMA(1,0,0)(1,0,0)[2] errors 
Box Cox transformation: lambda= 0 

Coefficients:
          ar1     sar1  intercept  pink_ind  lag1_log_SAR1  lag2_NPGO  lag1_NPGO
      -0.8622  -0.8583    11.9948   -1.0712         0.0211     0.2740     0.8406
s.e.   0.1909   0.1660     0.3683    0.1380         0.0564     0.0438     0.0822

sigma^2 = 0.03704:  log likelihood = 8.63
AIC=-1.26   AICc=-145.26   BIC=-0.62

Training set error measures:
                   ME     RMSE      MAE      MPE     MAPE      MASE       ACF1
Training set 3281.224 14791.52 11648.72 -1.07215 6.191168 0.1014961 -0.7177414
Series: . 
Regression with ARIMA(0,0,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      intercept  pink_ind  lag1_log_SAR1  lag2_NPGO  lag1_NPGO
        11.9743   -1.1456         0.0120     0.2011     0.8929
s.e.     0.4016    0.0951         0.0599     0.0545     0.0855

sigma^2 = 0.02986:  log likelihood = 6.68
AIC=-1.36   AICc=40.64   BIC=-0.18

Training set error measures:
                   ME     RMSE      MAE        MPE     MAPE      MASE       ACF1
Training set 5287.764 32720.75 18768.31 -0.6588648 9.731836 0.1907041 -0.2456846
Series: . 
Regression with ARIMA(0,0,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      intercept  pink_ind  lag1_log_SAR1  lag2_NPGO  lag1_NPGO
        12.9837   -1.1342         0.1534     0.2820     0.7001
s.e.     0.6878    0.1844         0.1043     0.1015     0.1505

sigma^2 = 0.0999:  log likelihood = 0.79
AIC=10.41   AICc=38.41   BIC=12.23

Training set error measures:
                   ME     RMSE     MAE       MPE     MAPE      MASE        ACF1
Training set 10437.08 77246.86 41868.5 -2.425846 16.99715 0.4253778 -0.03595432
Series: . 
Regression with ARIMA(0,0,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      intercept  pink_ind  lag1_log_SAR1  lag2_NPGO  lag1_NPGO
        15.5844   -0.4328         0.5613     0.3513     0.0360
s.e.     1.3206    0.3531         0.1973     0.2317     0.2656

sigma^2 = 0.4815:  log likelihood = -8.25
AIC=28.51   AICc=49.51   BIC=30.9

Training set error measures:
                   ME     RMSE      MAE       MPE     MAPE      MASE       ACF1
Training set 16115.13 139359.5 94655.63 -13.40343 43.18834 0.7656756 -0.1088019
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag2_NPGO  lag1_NPGO
       -0.7589         0.2316     0.1423     0.3301
s.e.    0.2389         0.1448     0.2289     0.2461

sigma^2 = 0.3851:  log likelihood = -7.87
AIC=25.75   AICc=37.75   BIC=27.74

Training set error measures:
                   ME     RMSE     MAE     MPE     MAPE      MASE        ACF1
Training set 49706.11 140457.7 94992.3 3.81621 32.68303 0.6862163 -0.05112375
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag2_NPGO  lag1_NPGO
       -0.7430         0.2628     0.1519     0.2794
s.e.    0.2275         0.1191     0.2197     0.2059

sigma^2 = 0.3422:  log likelihood = -8.16
AIC=26.32   AICc=36.32   BIC=28.74

Training set error measures:
                   ME     RMSE      MAE      MPE     MAPE      MASE         ACF1
Training set 41307.05 132581.8 90307.22 1.807594 32.50902 0.5496474 -0.005531777
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag2_NPGO  lag1_NPGO
       -0.7364         0.2347     0.2073     0.2731
s.e.    0.2215         0.1065     0.1939     0.2004

sigma^2 = 0.3131:  log likelihood = -8.51
AIC=27.02   AICc=35.59   BIC=29.84

Training set error measures:
                   ME   RMSE      MAE     MPE     MAPE      MASE        ACF1
Training set 40298.88 126948 85922.31 3.49573 32.08203 0.5271351 -0.01190667
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag2_NPGO  lag1_NPGO
       -0.7121         0.2363     0.2126     0.2488
s.e.    0.1851         0.1026     0.1858     0.1614

sigma^2 = 0.2829:  log likelihood = -8.67
AIC=27.34   AICc=34.84   BIC=30.54

Training set error measures:
                   ME     RMSE      MAE      MPE     MAPE      MASE        ACF1
Training set 37867.36 121637.8 80305.13 3.787277 30.64651 0.5270879 -0.02620781
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag2_NPGO  lag1_NPGO
       -0.9341         0.2827    -0.0238     0.3174
s.e.    0.1851         0.1160     0.1819     0.1829

sigma^2 = 0.3649:  log likelihood = -11.4
AIC=32.79   AICc=39.46   BIC=36.33

Training set error measures:
                   ME     RMSE      MAE      MPE     MAPE      MASE        ACF1
Training set 43911.41 137841.2 95344.67 4.982658 35.28384 0.6271043 -0.05476879
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag2_NPGO  lag1_NPGO
       -0.9425         0.2810    -0.0238     0.3185
s.e.    0.1761         0.1123     0.1764     0.1773

sigma^2 = 0.3357:  log likelihood = -11.67
AIC=33.34   AICc=39.34   BIC=37.2

Training set error measures:
                   ME     RMSE      MAE      MPE     MAPE     MASE        ACF1
Training set 39575.04 134138.3 90927.97 3.826038 34.06206 0.615134 -0.09194943
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag2_NPGO  lag1_NPGO
       -0.9214         0.2745    -0.0147     0.2999
s.e.    0.1569         0.1072     0.1691     0.1614

sigma^2 = 0.3116:  log likelihood = -11.93
AIC=33.86   AICc=39.31   BIC=38.03

Training set error measures:
                  ME     RMSE      MAE      MPE     MAPE      MASE        ACF1
Training set 32306.8 131447.8 90728.31 2.590159 33.07863 0.5717902 -0.06851705
Series: . 
Regression with ARIMA(0,0,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag2_PDO  lag1_PDO
       -0.2988        -1.3862    0.4994   -4.8298
s.e.    0.7052         0.0849    0.8623    0.7443

sigma^2 = 1.747:  log likelihood = -10.81
AIC=31.62   AICc=61.62   BIC=32.02

Training set error measures:
                   ME     RMSE      MAE       MPE     MAPE     MASE       ACF1
Training set -61043.9 310799.1 178050.5 -41.33743 95.45028 1.551365 0.03990726
Series: . 
Regression with ARIMA(0,0,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag2_PDO  lag1_PDO
       -0.6909        -1.4216    1.1005   -4.9759
s.e.    0.6565         0.0829    0.7471    0.7607

sigma^2 = 1.682:  log likelihood = -12.47
AIC=34.93   AICc=54.93   BIC=35.92

Training set error measures:
                    ME     RMSE      MAE       MPE     MAPE     MASE      ACF1
Training set -57911.81 294657.1 195902.2 -45.54447 107.3363 1.990555 0.2044282
Series: . 
Regression with ARIMA(0,0,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      intercept  pink_ind  lag1_log_SAR1  lag2_PDO  lag1_PDO
        15.9811   -0.6160         0.6172   -0.5921    0.2734
s.e.     1.8018    0.2737         0.2358    0.3026    0.6371

sigma^2 = 0.3284:  log likelihood = -5.16
AIC=22.31   AICc=50.31   BIC=24.13

Training set error measures:
                   ME     RMSE      MAE      MPE     MAPE      MASE        ACF1
Training set 7803.756 69205.73 58568.99 -8.37129 37.11919 0.5950523 -0.03237308
Series: . 
Regression with ARIMA(0,0,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      intercept  pink_ind  lag1_log_SAR1  lag2_PDO  lag1_PDO
        16.7362   -0.5608         0.7142   -0.6923    0.5772
s.e.     1.0328    0.2438         0.1395    0.2257    0.2471

sigma^2 = 0.281:  log likelihood = -5.29
AIC=22.59   AICc=43.59   BIC=24.97

Training set error measures:
                   ME     RMSE      MAE       MPE     MAPE      MASE         ACF1
Training set 5552.759 84824.57 67610.59 -7.864567 35.70873 0.5469065 -0.005638822
Series: . 
Regression with ARIMA(0,0,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      intercept  pink_ind  lag1_log_SAR1  lag2_PDO  lag1_PDO
        17.4874   -0.7762         0.7661   -0.5507    0.8260
s.e.     1.1608    0.2674         0.1626    0.2570    0.2645

sigma^2 = 0.3659:  log likelihood = -7.76
AIC=27.52   AICc=44.32   BIC=30.43

Training set error measures:
                   ME     RMSE      MAE       MPE     MAPE      MASE      ACF1
Training set 3869.658 101821.7 84607.76 -11.08727 42.64853 0.6111992 0.1587522
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag2_PDO  lag1_PDO
       -0.6356         0.4633   -0.1008    0.4997
s.e.    0.1358         0.1085    0.2207    0.2675

sigma^2 = 0.3302:  log likelihood = -7.95
AIC=25.89   AICc=35.89   BIC=28.32

Training set error measures:
                   ME     RMSE      MAE        MPE     MAPE      MASE        ACF1
Training set 19077.84 71923.55 43578.25 -0.8173965 28.56364 0.2652354 -0.09692394
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag2_PDO  lag1_PDO
       -0.6945         0.3886   -0.1099    0.3001
s.e.    0.1420         0.1077    0.2380    0.2615

sigma^2 = 0.3699:  log likelihood = -9.59
AIC=29.18   AICc=37.75   BIC=32.01

Training set error measures:
                   ME     RMSE      MAE       MPE     MAPE      MASE      ACF1
Training set 28617.93 87024.17 66185.88 0.7647625 36.14676 0.4060516 -0.178748
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag2_PDO  lag1_PDO
       -0.7145         0.3961   -0.0763    0.3172
s.e.    0.1329         0.1039    0.2226    0.2524

sigma^2 = 0.3396:  log likelihood = -9.95
AIC=29.9   AICc=37.4   BIC=33.09

Training set error measures:
                   ME     RMSE      MAE       MPE     MAPE      MASE       ACF1
Training set 22428.13 86304.49 63163.95 -1.052494 35.60261 0.4145807 -0.2218693
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag2_PDO  lag1_PDO
       -0.7816         0.4004   -0.1367    0.3941
s.e.    0.1370         0.1115    0.2363    0.2671

sigma^2 = 0.3812:  log likelihood = -11.72
AIC=33.45   AICc=40.12   BIC=36.99

Training set error measures:
                   ME     RMSE      MAE      MPE     MAPE      MASE       ACF1
Training set 28707.12 93219.25 71973.37 1.540995 39.62771 0.4733858 -0.2163904
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag2_PDO  lag1_PDO
       -0.7655         0.3941   -0.1155    0.3393
s.e.    0.1290         0.1078    0.2258    0.2323

sigma^2 = 0.3542:  log likelihood = -12.1
AIC=34.2   AICc=40.2   BIC=38.06

Training set error measures:
                   ME     RMSE      MAE     MPE     MAPE      MASE       ACF1
Training set 30584.55 94393.36 69747.27 2.75077 37.79934 0.4718452 -0.2339079
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag2_PDO  lag1_PDO
       -0.7733         0.4000   -0.1379    0.3354
s.e.    0.1224         0.1028    0.2052    0.2255

sigma^2 = 0.3286:  log likelihood = -12.38
AIC=34.76   AICc=40.22   BIC=38.93

Training set error measures:
                  ME     RMSE      MAE      MPE     MAPE      MASE       ACF1
Training set 33424.2 93225.06 70608.28 3.163543 36.73421 0.4449892 -0.2321161
Series: . 
Regression with ARIMA(0,0,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      intercept  lag1_log_SAR1  lag2_NPGO
        14.6205         0.4717     0.4789
s.e.     1.1905         0.1753     0.2335

sigma^2 = 0.4582:  log likelihood = -6.35
AIC=20.7   AICc=34.03   BIC=21.02

Training set error measures:
                   ME     RMSE      MAE       MPE     MAPE      MASE       ACF1
Training set 17748.99 113783.7 91141.15 -14.77848 51.31774 0.7941188 -0.4821734
Series: . 
Regression with ARIMA(0,0,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      intercept  lag1_log_SAR1  lag2_NPGO
        14.5837         0.4679     0.4735
s.e.     1.1167         0.1652     0.2199

sigma^2 = 0.3845:  log likelihood = -6.64
AIC=21.29   AICc=31.29   BIC=22.08

Training set error measures:
                   ME     RMSE      MAE       MPE    MAPE      MASE      ACF1
Training set 16486.59 107566.1 84291.29 -13.00808 46.5395 0.8564808 -0.497958
Series: . 
Regression with ARIMA(0,0,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      intercept  lag1_log_SAR1  lag2_NPGO
        14.7454         0.4901     0.5093
s.e.     1.0724         0.1590     0.2101

sigma^2 = 0.3506:  log likelihood = -7.16
AIC=22.33   AICc=30.33   BIC=23.54

Training set error measures:
                   ME     RMSE      MAE       MPE     MAPE      MASE       ACF1
Training set 20693.89 117647.6 92040.38 -12.97214 45.73147 0.9351167 -0.4870048
Series: . 
Regression with ARIMA(0,0,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      intercept  lag1_log_SAR1  lag2_NPGO
        15.2736         0.5544     0.4389
s.e.     1.1485         0.1724     0.2294

sigma^2 = 0.4154:  log likelihood = -9.03
AIC=26.05   AICc=32.72   BIC=27.64

Training set error measures:
                   ME     RMSE      MAE       MPE     MAPE      MASE       ACF1
Training set 23149.24 154340.3 110597.1 -16.70426 51.63265 0.8946274 -0.1877315
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      lag1_log_SAR1  lag2_NPGO
             0.1588     0.5893
s.e.         0.1781     0.2638

sigma^2 = 0.6043:  log likelihood = -11.73
AIC=29.47   AICc=32.9   BIC=30.66

Training set error measures:
                   ME   RMSE    MAE       MPE     MAPE     MASE       ACF1
Training set 43454.98 185966 142676 -2.413129 53.69426 1.030679 -0.3870771
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      lag1_log_SAR1  lag2_NPGO
             0.1830     0.6021
s.e.         0.1602     0.2521

sigma^2 = 0.5509:  log likelihood = -12.36
AIC=30.71   AICc=33.71   BIC=32.17

Training set error measures:
                   ME     RMSE      MAE       MPE     MAPE      MASE       ACF1
Training set 36788.23 183975.7 135206.9 -4.780421 52.50406 0.8229255 -0.3612322
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      lag1_log_SAR1  lag2_NPGO
             0.1608     0.6434
s.e.         0.1401     0.2110

sigma^2 = 0.5053:  log likelihood = -12.92
AIC=31.85   AICc=34.51   BIC=33.54

Training set error measures:
                   ME     RMSE      MAE       MPE     MAPE      MASE       ACF1
Training set 36044.15 177338.4 127115.6 -3.051009 49.93468 0.7798568 -0.3342366
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      lag1_log_SAR1  lag2_NPGO
             0.1592     0.6789
s.e.         0.1396     0.2068

sigma^2 = 0.495:  log likelihood = -13.86
AIC=33.73   AICc=36.13   BIC=35.64

Training set error measures:
                   ME     RMSE      MAE      MPE     MAPE      MASE       ACF1
Training set 28668.81 174847.7 121688.8 -8.57067 52.01299 0.7987125 -0.2947879
Series: . 
Regression with ARIMA(1,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
          ar1   drift  lag1_log_SAR1  lag2_NPGO
      -0.8647  0.2001         0.3596     0.1526
s.e.   0.1297  0.0748         0.0775     0.1252

sigma^2 = 0.3416:  log likelihood = -11.59
AIC=33.18   AICc=39.85   BIC=36.72

Training set error measures:
                    ME     RMSE      MAE       MPE     MAPE     MASE      ACF1
Training set -338.2715 127207.2 89737.56 -12.91573 42.42128 0.590225 0.1577966
Series: . 
Regression with ARIMA(1,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
          ar1   drift  lag1_log_SAR1  lag2_NPGO
      -0.8279  0.2124         0.3698     0.1308
s.e.   0.1272  0.0735         0.0769     0.1225

sigma^2 = 0.328:  log likelihood = -12.06
AIC=34.12   AICc=40.12   BIC=37.98

Training set error measures:
                    ME     RMSE      MAE       MPE     MAPE      MASE      ACF1
Training set -904.1313 124832.6 88321.25 -13.12479 43.54178 0.5974994 0.1929955
Series: . 
Regression with ARIMA(0,0,0)(0,1,0)[2] errors 
Box Cox transformation: lambda= 0 

Coefficients:
       drift  lag1_log_SAR1  lag2_NPGO
      0.1985         0.3575     0.1211
s.e.  0.0668         0.0704     0.1161

sigma^2 = 0.3268:  log likelihood = -12.09
AIC=32.19   AICc=35.82   BIC=35.28

Training set error measures:
                   ME     RMSE      MAE       MPE     MAPE      MASE       ACF1
Training set -24107.4 156240.5 106116.4 -12.44483 42.84954 0.6687694 0.05966511
Series: . 
Regression with ARIMA(0,0,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag2_NPGO  lag1_PDO
       -0.0758        -1.3751     0.2656   -4.2867
s.e.    0.6633         0.0821     0.4378    0.6496

sigma^2 = 1.74:  log likelihood = -10.8
AIC=31.59   AICc=61.59   BIC=31.99

Training set error measures:
                    ME     RMSE      MAE       MPE     MAPE     MASE      ACF1
Training set -52914.99 234537.5 148344.9 -33.86228 84.08531 1.292539 0.1550375
Series: . 
Regression with ARIMA(0,0,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag2_NPGO  lag1_PDO
       -0.4327        -1.4119     0.2993   -3.8644
s.e.    0.7192         0.0902     0.4951    0.6898

sigma^2 = 2.007:  log likelihood = -13.26
AIC=36.52   AICc=56.52   BIC=37.5

Training set error measures:
                    ME     RMSE      MAE       MPE     MAPE     MASE      ACF1
Training set -51534.93 228492.9 154592.8 -51.49034 105.8005 1.570812 0.3593672
Series: . 
Regression with ARIMA(0,0,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      intercept  pink_ind  lag1_log_SAR1  lag2_NPGO  lag1_PDO
        14.4511   -0.6352         0.4126     0.3874   -0.2599
s.e.     1.6499    0.2603         0.2142     0.1773    0.4857

sigma^2 = 0.3074:  log likelihood = -4.83
AIC=21.65   AICc=49.65   BIC=23.47

Training set error measures:
                   ME     RMSE      MAE       MPE     MAPE      MASE       ACF1
Training set 7603.932 76711.99 63099.77 -8.203267 33.11405 0.6410844 -0.3833386
Series: . 
Regression with ARIMA(0,0,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      intercept  pink_ind  lag1_log_SAR1  lag2_NPGO  lag1_PDO
        16.9671   -0.4919         0.7276     0.4470    0.5553
s.e.     1.1536    0.2798         0.1578     0.1957    0.2777

sigma^2 = 0.3537:  log likelihood = -6.56
AIC=25.12   AICc=46.12   BIC=27.5

Training set error measures:
                  ME     RMSE      MAE       MPE     MAPE      MASE      ACF1
Training set 12878.9 95206.25 74580.59 -10.75591 40.67701 0.6032873 -0.296785
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag2_NPGO  lag1_PDO
       -0.5673         0.4971     0.3231    0.7545
s.e.    0.1447         0.1277     0.1825    0.2605

sigma^2 = 0.2543:  log likelihood = -5.59
AIC=21.18   AICc=33.18   BIC=23.17

Training set error measures:
                    ME     RMSE     MAE      MPE     MAPE      MASE       ACF1
Training set -2261.173 90312.57 58493.5 1.101467 26.50938 0.4225521 -0.0279658
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag2_NPGO  lag1_PDO
       -0.5049         0.3785     0.3069    0.5529
s.e.    0.1498         0.1135     0.1952    0.2472

sigma^2 = 0.2786:  log likelihood = -6.93
AIC=23.85   AICc=33.85   BIC=26.28

Training set error measures:
                   ME     RMSE      MAE      MPE     MAPE      MASE       ACF1
Training set 18495.71 65877.24 49721.09 4.839352 27.62017 0.3026233 -0.4253366
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag2_NPGO  lag1_PDO
       -0.5067         0.3071     0.3914    0.4438
s.e.    0.1537         0.1021     0.1890    0.2388

sigma^2 = 0.2827:  log likelihood = -7.84
AIC=25.69   AICc=34.26   BIC=28.51

Training set error measures:
                   ME     RMSE      MAE      MPE     MAPE      MASE       ACF1
Training set 25134.14 70160.23 54762.21 7.201424 29.77926 0.3359673 -0.5325099
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag2_NPGO  lag1_PDO
       -0.5145         0.3098     0.3959    0.4573
s.e.    0.1461         0.0983     0.1820    0.2260

sigma^2 = 0.256:  log likelihood = -7.97
AIC=25.94   AICc=33.44   BIC=29.14

Training set error measures:
                   ME     RMSE      MAE      MPE     MAPE      MASE       ACF1
Training set 21966.59 67019.19 51717.46 5.941247 28.67348 0.3394509 -0.5477317
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag2_NPGO  lag1_PDO
       -0.7263         0.3715     0.1298    0.4275
s.e.    0.1570         0.1183     0.1951    0.2782

sigma^2 = 0.3785:  log likelihood = -11.67
AIC=33.34   AICc=40.01   BIC=36.88

Training set error measures:
                  ME     RMSE      MAE      MPE    MAPE      MASE       ACF1
Training set 26388.5 100893.9 73053.27 5.574825 35.9792 0.4804885 -0.2535356
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag2_NPGO  lag1_PDO
       -0.7164         0.3684     0.1155    0.3706
s.e.    0.1516         0.1152     0.1877    0.2416

sigma^2 = 0.3517:  log likelihood = -12.04
AIC=34.08   AICc=40.08   BIC=37.95

Training set error measures:
                   ME     RMSE      MAE      MPE     MAPE      MASE      ACF1
Training set 28862.72 99536.42 70124.31 6.078726 34.68653 0.4743958 -0.276017
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_log_SAR1  lag2_NPGO  lag1_PDO
       -0.7419         0.3855     0.0971     0.348
s.e.    0.1425         0.1093     0.1814     0.234

sigma^2 = 0.3317:  log likelihood = -12.46
AIC=34.92   AICc=40.38   BIC=39.09

Training set error measures:
                   ME     RMSE      MAE      MPE     MAPE      MASE       ACF1
Training set 34991.46 106874.7 79226.83 6.821568 35.25921 0.4993053 -0.2691993
Series: . 
Regression with ARIMA(0,0,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      lag1_log_SAR1  lag1_PDO
            -1.3635   -4.5165
s.e.         0.0689    0.5335

sigma^2 = 1.221:  log likelihood = -11
AIC=28   AICc=34   BIC=28.24

Training set error measures:
                    ME   RMSE      MAE      MPE     MAPE     MASE      ACF1
Training set -39705.64 217081 146726.3 -39.9175 92.95288 1.278436 0.1163397
Series: . 
Regression with ARIMA(0,0,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      intercept  lag1_log_SAR1  lag1_PDO
        13.3229         0.3220   -0.7467
s.e.     2.8085         0.3603    0.7565

sigma^2 = 0.5257:  log likelihood = -8.05
AIC=24.1   AICc=34.1   BIC=24.89

Training set error measures:
                   ME     RMSE      MAE       MPE     MAPE      MASE       ACF1
Training set 17777.83 128639.9 98233.25 -18.54234 56.47848 0.9981446 -0.5077593
Series: . 
Regression with ARIMA(0,0,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      intercept  lag1_log_SAR1  lag1_PDO
        15.1212         0.5490   -0.3714
s.e.     2.5765         0.3321    0.7327

sigma^2 = 0.5425:  log likelihood = -9.35
AIC=26.7   AICc=34.7   BIC=27.91

Training set error measures:
                   ME     RMSE      MAE       MPE     MAPE     MASE       ACF1
Training set 23962.08 179925.8 125751.5 -22.19296 60.57453 1.277616 -0.3940918
Series: . 
Regression with ARIMA(0,0,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      intercept  lag1_log_SAR1  lag1_PDO
        17.3305         0.8181    0.2842
s.e.     1.4584         0.2083    0.3733

sigma^2 = 0.5259:  log likelihood = -10.32
AIC=28.64   AICc=35.31   BIC=30.24

Training set error measures:
                  ME     RMSE      MAE       MPE     MAPE     MASE       ACF1
Training set 23640.7 195068.7 129703.3 -24.37685 61.91187 1.049178 -0.2044839
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      lag1_log_SAR1  lag1_PDO
             0.4631    0.4089
s.e.         0.2396    0.5162

sigma^2 = 0.8311:  log likelihood = -13.49
AIC=32.97   AICc=36.4   BIC=34.17

Training set error measures:
                   ME     RMSE      MAE       MPE     MAPE     MASE       ACF1
Training set 16621.52 198766.2 164430.7 -13.55758 60.00296 1.187833 -0.8465073
Series: . 
Regression with ARIMA(0,0,0)(0,1,0)[2] errors 
Box Cox transformation: lambda= 0 

Coefficients:
      lag1_log_SAR1  lag1_PDO
             0.5039    0.3599
s.e.         0.0997    0.2185

sigma^2 = 0.4202:  log likelihood = -9.74
AIC=25.47   AICc=28.9   BIC=26.66

Training set error measures:
                  ME     RMSE      MAE      MPE     MAPE      MASE      ACF1
Training set 43150.7 87382.85 62780.15 11.33379 32.38427 0.3821062 0.2880038
Series: . 
Regression with ARIMA(0,0,0)(0,1,0)[2] errors 
Box Cox transformation: lambda= 0 

Coefficients:
      lag1_log_SAR1  lag1_PDO
             0.3870    0.1449
s.e.         0.0894    0.2111

sigma^2 = 0.5123:  log likelihood = -11.92
AIC=29.84   AICc=32.84   BIC=31.29

Training set error measures:
                   ME     RMSE     MAE      MPE     MAPE      MASE      ACF1
Training set 61432.42 109413.6 83873.8 13.39909 38.54158 0.5145674 0.1573724
Series: . 
Regression with ARIMA(0,0,0)(0,1,0)[2] errors 
Box Cox transformation: lambda= 0 

Coefficients:
       drift  lag1_log_SAR1  lag1_PDO
      0.1768         0.4192    0.1422
s.e.  0.0771         0.0730    0.1633

sigma^2 = 0.373:  log likelihood = -10.33
AIC=28.66   AICc=33.66   BIC=30.92

Training set error measures:
                    ME     RMSE     MAE       MPE    MAPE      MASE      ACF1
Training set -14620.64 125692.3 92346.8 -14.10441 45.3792 0.6061242 0.1331245
Series: . 
Regression with ARIMA(0,0,0)(0,1,0)[2] errors 
Box Cox transformation: lambda= 0 

Coefficients:
       drift  lag1_log_SAR1  lag1_PDO
      0.1927         0.4194    0.1438
s.e.  0.0736         0.0721    0.1612

sigma^2 = 0.3558:  log likelihood = -10.94
AIC=29.89   AICc=34.33   BIC=32.44

Training set error measures:
                    ME   RMSE      MAE      MPE    MAPE      MASE      ACF1
Training set -14092.78 131538 96686.35 -14.1617 45.8357 0.6359288 0.1416106
Series: . 
Regression with ARIMA(0,0,0)(0,1,0)[2] errors 
Box Cox transformation: lambda= 0 

Coefficients:
       drift  lag1_log_SAR1  lag1_PDO
      0.2106         0.4212    0.1147
s.e.  0.0712         0.0719    0.1582

sigma^2 = 0.3484:  log likelihood = -11.7
AIC=31.4   AICc=35.4   BIC=34.24

Training set error measures:
                    ME     RMSE      MAE       MPE     MAPE      MASE      ACF1
Training set -17339.67 139625.6 97479.83 -14.56302 46.68839 0.6594579 0.1735344
Series: . 
Regression with ARIMA(0,0,0)(0,1,0)[2] errors 
Box Cox transformation: lambda= 0 

Coefficients:
       drift  lag1_log_SAR1  lag1_PDO
      0.2002         0.4141    0.1438
s.e.  0.0677         0.0696    0.1477

sigma^2 = 0.3295:  log likelihood = -12.16
AIC=32.32   AICc=35.95   BIC=35.41

Training set error measures:
                    ME     RMSE      MAE       MPE     MAPE      MASE      ACF1
Training set -24557.19 136759.8 102504.6 -13.73937 45.48019 0.6460072 0.1071716
Series: . 
Regression with ARIMA(0,0,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      intercept  pink_ind  lag1_NPGO  lag1_PDO
        11.8733   -1.1885     0.9445   -0.1881
s.e.     0.1591    0.1491     0.1212    0.2039

sigma^2 = 0.06491:  log likelihood = 2.36
AIC=5.28   AICc=35.28   BIC=5.68

Training set error measures:
                   ME     RMSE      MAE       MPE     MAPE     MASE       ACF1
Training set 5564.969 43659.14 32897.83 -1.617791 15.60242 0.286641 0.07790114
Series: . 
Regression with ARIMA(0,0,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      intercept  pink_ind  lag1_NPGO  lag1_PDO
        11.9237   -1.2401     0.9605   -0.1151
s.e.     0.1491    0.1381     0.1193    0.1878

sigma^2 = 0.05777:  log likelihood = 2.71
AIC=4.59   AICc=24.59   BIC=5.58

Training set error measures:
                   ME  RMSE      MAE       MPE     MAPE      MASE        ACF1
Training set 5873.569 46773 34263.83 -1.587919 15.70193 0.3481536 -0.06561069
Series: . 
Regression with ARIMA(0,0,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      intercept  pink_ind  lag1_NPGO  lag1_PDO
        12.0498   -1.4118     0.9264   -0.2120
s.e.     0.2622    0.2382     0.2128    0.3336

sigma^2 = 0.1707:  log likelihood = -2.8
AIC=15.59   AICc=30.59   BIC=17.11

Training set error measures:
                   ME     RMSE      MAE       MPE     MAPE      MASE        ACF1
Training set 13373.84 111259.9 58795.77 -4.642584 22.42453 0.5973564 -0.03602239
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_NPGO  lag1_PDO
       -1.1635     0.8090    0.5056
s.e.    0.2214     0.2605    0.3099

sigma^2 = 0.3723:  log likelihood = -7.47
AIC=22.93   AICc=30.93   BIC=24.14

Training set error measures:
                   ME     RMSE      MAE      MPE     MAPE      MASE       ACF1
Training set 71751.25 151962.3 104452.3 7.769751 35.02222 0.8449213 -0.1156412
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_NPGO  lag1_PDO
       -1.1302     0.8748    0.4939
s.e.    0.2295     0.2672    0.3230

sigma^2 = 0.3897:  log likelihood = -8.67
AIC=25.35   AICc=32.01   BIC=26.94

Training set error measures:
                   ME     RMSE      MAE        MPE    MAPE      MASE       ACF1
Training set 42236.88 180639.8 126135.5 -0.4114352 38.6365 0.9111921 -0.3082373
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_NPGO  lag1_PDO
       -1.1522     0.7516    0.4596
s.e.    0.2598     0.2954    0.3656

sigma^2 = 0.4851:  log likelihood = -10.96
AIC=29.92   AICc=35.64   BIC=31.86

Training set error measures:
                   ME     RMSE      MAE      MPE     MAPE      MASE        ACF1
Training set 15093.36 187027.4 132705.6 -17.3744 54.23814 0.8077019 -0.05220916
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_NPGO  lag1_PDO
       -1.1499     0.7039    0.3672
s.e.    0.2544     0.2812    0.3330

sigma^2 = 0.4538:  log likelihood = -11.61
AIC=31.21   AICc=36.21   BIC=33.47

Training set error measures:
                   ME     RMSE      MAE       MPE     MAPE      MASE        ACF1
Training set 18479.21 183482.5 126607.1 -13.66421 52.54034 0.7767368 -0.06587427
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_NPGO  lag1_PDO
       -0.9821     0.5133    0.2075
s.e.    0.1973     0.2139    0.2963

sigma^2 = 0.4469:  log likelihood = -12.54
AIC=33.08   AICc=37.52   BIC=35.63

Training set error measures:
                   ME     RMSE      MAE       MPE     MAPE      MASE        ACF1
Training set 20154.93 177469.3 120990.6 -11.40399 53.43693 0.7941293 -0.06045579
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_NPGO  lag1_PDO
       -1.0573     0.5474    0.2782
s.e.    0.1938     0.2177    0.2988

sigma^2 = 0.4612:  log likelihood = -13.81
AIC=35.61   AICc=39.61   BIC=38.45

Training set error measures:
                   ME     RMSE      MAE       MPE     MAPE      MASE        ACF1
Training set 25879.22 181837.3 123949.6 -7.106354 53.50035 0.8152455 -0.06409897
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_NPGO  lag1_PDO
       -1.0349     0.5328    0.2281
s.e.    0.1792     0.2084    0.2597

sigma^2 = 0.4297:  log likelihood = -14.28
AIC=36.57   AICc=40.21   BIC=39.66

Training set error measures:
                   ME     RMSE      MAE       MPE     MAPE      MASE        ACF1
Training set 27868.72 175488.6 117557.1 -5.671838 51.03973 0.7952818 -0.05932447
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag1_NPGO  lag1_PDO
       -1.0441     0.5426    0.2259
s.e.    0.1602     0.1890    0.2516

sigma^2 = 0.3995:  log likelihood = -14.67
AIC=37.34   AICc=40.68   BIC=40.68

Training set error measures:
                   ME     RMSE      MAE       MPE     MAPE      MASE        ACF1
Training set 28941.33 170734.9 113248.1 -4.861987 48.45115 0.7137149 -0.05827991
Series: . 
Regression with ARIMA(0,0,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      intercept  pink_ind  lag2_NPGO  lag1_NPGO
        11.8890   -1.1264     0.2056     0.9209
s.e.     0.0624    0.0873     0.0527     0.0533

sigma^2 = 0.02476:  log likelihood = 6.21
AIC=-2.43   AICc=27.57   BIC=-2.03

Training set error measures:
                   ME     RMSE      MAE        MPE     MAPE     MASE        ACF1
Training set 5521.859 33540.17 18527.12 -0.6116831 9.225256 0.161428 -0.02638776
Series: . 
Regression with ARIMA(0,0,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      intercept  pink_ind  lag2_NPGO  lag1_NPGO
        11.8947   -1.1528     0.2012     0.9062
s.e.     0.0646    0.0881     0.0546     0.0541

sigma^2 = 0.02399:  log likelihood = 6.66
AIC=-3.32   AICc=16.68   BIC=-2.33

Training set error measures:
                   ME     RMSE      MAE        MPE     MAPE      MASE       ACF1
Training set 5348.989 33645.97 18432.25 -0.6593162 9.465679 0.1872894 -0.2667025
Series: . 
Regression with ARIMA(0,0,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      intercept  pink_ind  lag2_NPGO  lag1_NPGO
        11.9884   -1.2466     0.3041     0.8604
s.e.     0.1339    0.1851     0.1107     0.1143

sigma^2 = 0.1012:  log likelihood = -0.18
AIC=10.37   AICc=25.37   BIC=11.88

Training set error measures:
                   ME     RMSE      MAE       MPE   MAPE     MASE        ACF1
Training set 12241.56 91874.11 44743.57 -2.802921 16.481 0.454588 -0.05646057
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag2_NPGO  lag1_NPGO
       -0.8336     0.2154     0.4628
s.e.    0.2788     0.2481     0.2557

sigma^2 = 0.4384:  log likelihood = -8.28
AIC=24.56   AICc=32.56   BIC=25.77

Training set error measures:
                   ME   RMSE     MAE     MPE     MAPE      MASE      ACF1
Training set 75241.13 155230 97065.6 9.30864 33.37343 0.7851701 0.1746257
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag2_NPGO  lag1_NPGO
       -0.7614     0.2871     0.4781
s.e.    0.2652     0.2335     0.2532

sigma^2 = 0.4154:  log likelihood = -9.02
AIC=26.05   AICc=32.72   BIC=27.64

Training set error measures:
                   ME   RMSE      MAE      MPE     MAPE      MASE       ACF1
Training set 52673.95 162964 111195.8 4.439343 36.09832 0.8032692 -0.1581367
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag2_NPGO  lag1_NPGO
       -0.6964     0.4081     0.3504
s.e.    0.2686     0.2213     0.2411

sigma^2 = 0.4277:  log likelihood = -10.21
AIC=28.41   AICc=34.13   BIC=30.35

Training set error measures:
                   ME     RMSE      MAE       MPE     MAPE      MASE        ACF1
Training set 35271.85 148992.2 101393.8 -2.964593 40.26085 0.6171247 -0.01731898
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag2_NPGO  lag1_NPGO
       -0.6973     0.3923     0.3583
s.e.    0.2587     0.2049     0.2305

sigma^2 = 0.3871:  log likelihood = -10.57
AIC=29.14   AICc=34.14   BIC=31.4

Training set error measures:
                   ME   RMSE      MAE       MPE     MAPE      MASE        ACF1
Training set 31295.31 144839 96484.89 -4.150003 38.57642 0.5919367 0.007346216
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag2_NPGO  lag1_NPGO
       -0.6562     0.4032     0.3186
s.e.    0.2155     0.1953     0.1862

sigma^2 = 0.3546:  log likelihood = -10.92
AIC=29.84   AICc=34.28   BIC=32.39

Training set error measures:
                   ME     RMSE      MAE       MPE     MAPE      MASE         ACF1
Training set 29587.33 138784.3 90801.02 -3.008671 37.14574 0.5959784 -0.004463171
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag2_NPGO  lag1_NPGO
       -0.9137     0.1604     0.4191
s.e.    0.2184     0.1955     0.2104

sigma^2 = 0.4669:  log likelihood = -13.9
AIC=35.8   AICc=39.8   BIC=38.63

Training set error measures:
                   ME     RMSE      MAE       MPE     MAPE      MASE        ACF1
Training set 34993.03 165858.5 108626.8 -3.070926 45.04264 0.7144639 -0.04155968
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag2_NPGO  lag1_NPGO
       -0.9168     0.1600     0.4193
s.e.    0.2074     0.1892     0.2037

sigma^2 = 0.4312:  log likelihood = -14.31
AIC=36.62   AICc=40.26   BIC=39.71

Training set error measures:
                  ME     RMSE      MAE       MPE     MAPE      MASE        ACF1
Training set 32316.2 161074.2 102607.7 -3.179883 42.69566 0.6941484 -0.06057497
Series: . 
Regression with ARIMA(0,1,0) errors 
Box Cox transformation: lambda= 0 

Coefficients:
      pink_ind  lag2_NPGO  lag1_NPGO
       -0.9302     0.1569     0.4329
s.e.    0.1846     0.1828     0.1799

sigma^2 = 0.401:  log likelihood = -14.7
AIC=37.41   AICc=40.74   BIC=40.74

Training set error measures:
                  ME     RMSE     MAE       MPE     MAPE      MASE        ACF1
Training set 33549.8 157241.9 99805.1 -2.456943 40.75958 0.6289942 -0.05874416