Merge pull request #12 from USEPA/develop

CRAN Version 0.5.0

.Rbuildignore

@@ -10,3 +10,5 @@
 ^_pkgdown\.yml$
 ^docs$
 ^pkgdown$
+^vignettes/articles$
+^CRAN-SUBMISSION$

.github/workflows/check-standard.yaml → .github/workflows/R-CMD-check.yaml

@@ -1,4 +1,4 @@
-# Workflow derived from https://github.com/r-lib/actions/tree/master/examples
+# Workflow derived from https://github.com/r-lib/actions/tree/v2/examples
 # Need help debugging build failures? Start at https://github.com/r-lib/actions#where-to-find-help
 on:
   push:
@@ -18,7 +18,7 @@ jobs:
       fail-fast: false
       matrix:
         config:
-          - {os: macOS-latest,   r: 'release'}
+          - {os: macos-latest,   r: 'release'}
           - {os: windows-latest, r: 'release'}
           - {os: ubuntu-latest,   r: 'devel', http-user-agent: 'release'}
           - {os: ubuntu-latest,   r: 'release'}
@@ -29,9 +29,9 @@ jobs:
       R_KEEP_PKG_SOURCE: yes
 
     steps:
-      - uses: actions/checkout@v2
+      - uses: actions/checkout@v3
 
-      - uses: r-lib/actions/setup-pandoc@v1
+      - uses: r-lib/actions/setup-pandoc@v2
 
       - uses: r-lib/actions/setup-r@v2
         with:
@@ -41,22 +41,9 @@ jobs:
 
       - uses: r-lib/actions/setup-r-dependencies@v2
         with:
-          extra-packages: rcmdcheck
-
-      - uses: r-lib/actions/setup-tinytex@v1
+          extra-packages: any::rcmdcheck
+          needs: check
 
       - uses: r-lib/actions/check-r-package@v2
         with:
-          build_args: '"--compact-vignettes=both"'
-
-      - name: Show testthat output
-        if: always()
-        run: find check -name 'testthat.Rout*' -exec cat '{}' \; || true
-        shell: bash
-
-      - name: Upload check results
-        if: always()
-        uses: actions/upload-artifact@main
-        with:
-          name: ${{ runner.os }}-r${{ matrix.config.r }}-results
-          path: check
+          upload-snapshots: true

DESCRIPTION

@@ -1,6 +1,6 @@
 Package: spmodel
 Title: Spatial Statistical Modeling and Prediction
-Version: 0.4.0
+Version: 0.5.0
 Authors@R: c(
     person(given = "Michael",
            family = "Dumelle",
@@ -21,7 +21,7 @@ License: GPL-3
 Encoding: UTF-8
 LazyData: true
 Roxygen: list(markdown = TRUE)
-RoxygenNote: 7.1.2
+RoxygenNote: 7.2.3
 Depends: 
     R (>= 3.5.0)
 Imports: 

NEWS.md

@@ -1,3 +1,27 @@
+# spmodel 0.5.0
+
+## Minor updates
+
+* Predictions can now be made for prediction locations whose random effect levels are not present in the observed data
+    * When this occurs, the random-effect covariance between the observed data and these prediction locations is assumed to be zero.
+* The default for `local = TRUE` in `splm()` and `spglm()` now uses the `kmeans` assignment method with group sizes approximately equal to 100.
+    * Previously, the `random` assignment method was used with group sizes approximately equal to 50.
+* The default for `local = TRUE` in `predict()` and `augment()` now uses 100 local neighbors.
+    * Previously, 50 local neighbors were used.
+* Moved the "A Detailed Guide to `spmodel`" and "Technical Details" vignettes to the package website.
+* Added a "Spatial Generalized Linear Models in `spmodel`" vignette to the package website.
+* Changed name of "An Overview of Basic Features in `spmodel`" vignette to "An Introduction to `spmodel`" and changed output type from PDF to HTML.
+* Other minor vignette updates.
+* Minor documentation updates.
+
+## Bug fixes
+
+* Fixed a bug that occurred with prediction for success/failure binomial data (e.g., Bernoulli data) when `local` in `predict()` was `TRUE`.
+* Fixed a bug that could affect simulating data using `sprbinom()` when the `size` argument was different from `1`.
+* Fixed a bug that could cause local prediction to fail when only one level of a random effect was present in the prediction site's local neighborhood.
+* Fixed a bug that could cause an error when local estimation was used for the `"sv-wls"` estimation method.
+* Fixed a bug that caused undesirable behavior from `tidy()` when `conf.level` was less than zero or greater than one.
+
 # spmodel 0.4.0
 
 ## Major updates

R/augment.R

@@ -49,20 +49,20 @@
 #'
 #' @return When augmenting the original data set, a tibble with additional columns
 #'   \itemize{
-#'     \item{\code{.fitted}}{ Fitted value}
-#'     \item{\code{.resid}}{ Response residual (the difference between observed and fitted values)}
-#'     \item{\code{.hat}}{ Leverage (diagonal of the hat matrix)}
-#'     \item{\code{.cooksd}}{ Cook's distance}
-#'     \item{\code{.std.resid}}{ Standardized residuals}
-#'     \item{\code{.se.fit}}{ Standard error of the fitted value.}
+#'     \item \code{.fitted} Fitted value
+#'     \item \code{.resid} Response residual (the difference between observed and fitted values)
+#'     \item \code{.hat} Leverage (diagonal of the hat matrix)
+#'     \item \code{.cooksd} Cook's distance
+#'     \item \code{.std.resid} Standardized residuals
+#'     \item \code{.se.fit} Standard error of the fitted value.
 #'   }
 #'
 #'   When augmenting a new data set, a tibble with additional columns
 #'   \itemize{
-#'     \item{\code{.fitted}}{ Predicted (or fitted) value}
-#'     \item{\code{.lower}}{ Lower bound on interval}
-#'     \item{\code{.upper}}{ Upper bound on interval}
-#'     \item{\code{.se.fit}}{ Standard error of the predicted (or fitted) value}
+#'     \item \code{.fitted} Predicted (or fitted) value
+#'     \item \code{.lower} Lower bound on interval
+#'     \item \code{.upper} Upper bound on interval
+#'     \item \code{.se.fit} Standard error of the predicted (or fitted) value
 #'   }
 #'
 #' @name augment.spmodel

R/cov_estimate_sv.R

@@ -8,7 +8,7 @@
 #' @param optim_dotlist The optim dotlist
 #' @param esv_dotlist The empirical semivariogram dotlist
 #'
-#' @return
+#' @return Covariance parameter estimates
 #'
 #' @noRd
 cov_estimate_sv <- function(data_object, formula, spcov_initial, estmethod,

R/cov_initial_search.R

@@ -78,6 +78,8 @@ cov_initial_search.exponential <- function(spcov_initial_NA, estmethod, data_obj
       ))
     }, SIMPLIFY = FALSE)
     esv_vals <- do.call("rbind", esv_vals)
+    esv_vals <- esv_vals[esv_vals$np > 0, , drop = FALSE]
+    esv_vals$bins <- droplevels(esv_vals$bins)
     esv_val <- data.frame(
       bins = levels(esv_vals$bins),
       dist = tapply(esv_vals$dist, esv_vals$bins, function(x) mean(x, na.rm = TRUE)),
@@ -365,6 +367,8 @@ cov_initial_search.none <- function(spcov_initial_NA, estmethod, data_object,
       ))
     }, SIMPLIFY = FALSE)
     esv_vals <- do.call("rbind", esv_vals)
+    esv_vals <- esv_vals[esv_vals$np > 0, , drop = FALSE]
+    esv_vals$bins <- droplevels(esv_vals$bins)
     esv_val <- data.frame(
       bins = levels(esv_vals$bins),
       dist = tapply(esv_vals$dist, esv_vals$bins, function(x) mean(x, na.rm = TRUE)),
@@ -637,6 +641,8 @@ cov_initial_search.matern <- function(spcov_initial_NA, estmethod, data_object,
       ))
     }, SIMPLIFY = FALSE)
     esv_vals <- do.call("rbind", esv_vals)
+    esv_vals <- esv_vals[esv_vals$np > 0, , drop = FALSE]
+    esv_vals$bins <- droplevels(esv_vals$bins)
     esv_val <- data.frame(
       bins = levels(esv_vals$bins),
       dist = tapply(esv_vals$dist, esv_vals$bins, function(x) mean(x, na.rm = TRUE)),

R/data.R

@@ -5,21 +5,21 @@
 #' @format An \code{sf} object with 365 rows and 10 columns:
 #'
 #' \itemize{
-#'   \item{sample: }{A factor with a sample identifier. Some samples were
+#'   \item sample: A factor with a sample identifier. Some samples were
 #'     replicated in the field or laboratory. As a result, there are 318 unique
-#'     sample identifiers.}
-#'   \item{field_dup: }{A factor representing field duplicate. Takes values \code{1}
-#'     and \code{2}.}
-#'   \item{lab_rep: }{A factor representing laboratory replicate. Takes values \code{1}
-#'     and \code{2}.}
-#'   \item{year: }{A factor representing year. Takes values \code{2001} and \code{2006}.}
-#'   \item{sideroad: }{A factor representing direction relative to the haul road.
+#'     sample identifiers.
+#'   \item field_dup: A factor representing field duplicate. Takes values \code{1}
+#'     and \code{2}.
+#'   \item lab_rep: A factor representing laboratory replicate. Takes values \code{1}
+#'     and \code{2}.
+#'   \item year: A factor representing year. Takes values \code{2001} and \code{2006}.
+#'   \item sideroad: A factor representing direction relative to the haul road.
 #'     Takes values \code{N} (north of the haul road) and \code{S} (south
-#'     of the haul road).}
-#'   \item{log_dist2road: }{The log of distance (in meters) to the haul road.}
-#'   \item{log_Zn: }{The log of zinc concentration in moss tissue (mg/kg).}
-#'   \item{geometry: }{\code{POINT} geometry representing coordinates in an Alaska
-#'     Albers projection (EPSG: 3338).}
+#'     of the haul road).
+#'   \item log_dist2road: The log of distance (in meters) to the haul road.
+#'   \item log_Zn: The log of zinc concentration in moss tissue (mg/kg).
+#'   \item geometry: \code{POINT} geometry representing coordinates in an Alaska
+#'     Albers projection (EPSG: 3338).
 #' }
 #' @source Data were obtained from Peter Neitlich and Linda Hasselbach of the National
 #'   Park Service.  Data were used in the publications listed in References.
@@ -43,13 +43,13 @@
 #'
 #' @format A \code{tibble} with 30 rows and 5 columns:
 #' \itemize{
-#'   \item{water: }{A factor representing whether water was added. Takes values
-#'     \code{N} (no water added) and \code{Y} (water added).}
-#'   \item{tarp: }{A factor representing tarp cover. Takes values \code{clear}
-#'     (a clear tarp), \code{shade} (a shade tarp), and \code{none} (no tarp).}
-#'   \item{z: }{The percentage of nitrogen.}
-#'   \item{x: }{The x-coordinate.}
-#'   \item{y: }{The y-coordinate.}
+#'   \item water: A factor representing whether water was added. Takes values
+#'     \code{N} (no water added) and \code{Y} (water added).
+#'   \item tarp: A factor representing tarp cover. Takes values \code{clear}
+#'     (a clear tarp), \code{shade} (a shade tarp), and \code{none} (no tarp).
+#'   \item z: The percentage of nitrogen.
+#'   \item x: The x-coordinate.
+#'   \item y: The y-coordinate.
 #' }
 #' @source These data were provided by Elizabeth Lenart of the Alaska Department
 #'   of Fish and Game.  The data were used in the publication listed in References.
@@ -63,10 +63,10 @@
 #' @description Estimated harbor-seal trends from abundance data in southeast Alaska, USA.
 #'
 #' @format A \code{sf} object with 62 rows and 2 columns:
-#' \describe{
-#'   \item{log_trend: }{The log of the estimated harbor-seal trends from abundance data.}
-#'   \item{geometry: }{\code{POLYGON} geometry representing polygons in an Alaska
-#'     Albers projection (EPSG: 3338).}
+#' \itemize{
+#'   \item log_trend: The log of the estimated harbor-seal trends from abundance data.
+#'   \item geometry: \code{POLYGON} geometry representing polygons in an Alaska
+#'     Albers projection (EPSG: 3338).
 #' }
 #' @source These data were collected by the Polar Ecosystem Program of the Marine
 #'   Mammal Laboratory of the Alaska Fisheries Science Center of NOAA Fisheries.
@@ -82,10 +82,10 @@
 #' @description Sulfate atmospheric deposition in the conterminous USA.
 #'
 #' @format An \code{sf} object with 197 rows and 2 columns.
-#' \describe{
-#'   \item{sulfate: }{Total wet deposition sulfate in kilograms per hectare.}
-#'   \item{geometry: }{\code{POINT} geometry representing coordinates in a
-#'     Conus Albers projection (EPSG: 5070).}
+#' \itemize{
+#'   \item sulfate: Total wet deposition sulfate in kilograms per hectare.
+#'   \item geometry: \code{POINT} geometry representing coordinates in a
+#'     Conus Albers projection (EPSG: 5070).
 #' }
 #' @source
 #' These data were used in the publication listed in References. Data were downloaded from the
@@ -101,9 +101,9 @@
 #' @description Locations at which to predict sulfate atmospheric deposition in the conterminous USA.
 #'
 #' @format An \code{sf} object with 197 rows and 1 column.
-#' \describe{
-#'   \item{geometry: }{\code{POINT} geometry representing coordinates in a
-#'     Conus Albers projection (EPSG: 5070).}
+#' \itemize{
+#'   \item geometry: \code{POINT} geometry representing coordinates in a
+#'     Conus Albers projection (EPSG: 5070).
 #' }
 #' @source
 #' These data were used in the publication listed in References. Data were downloaded from
@@ -119,14 +119,14 @@
 #' @description Moose counts and presence in Alaska, USA.
 #'
 #' @format An \code{sf} object with 218 rows and 5 columns.
-#' \describe{
-#'   \item{elev: }{The elevation.}
-#'   \item{strat: }{A factor representing strata (used for sampling). Can take values \code{L} and \code{M}.}
-#'   \item{count: }{The count (number) of moose observed.}
-#'   \item{presence: }{A binary factor representing whether no moose were observed (value \code{0}) or at least one moose was observed
-#'     (value \code{1}).}
-#'   \item{geometry: }{\code{POINT} geometry representing coordinates in an Alaska
-#'     Albers projection (EPSG: 3338).}
+#' \itemize{
+#'   \item elev: The elevation.
+#'   \item strat: A factor representing strata (used for sampling). Can take values \code{L} and \code{M}.
+#'   \item count: The count (number) of moose observed.
+#'   \item presence: A binary factor representing whether no moose were observed (value \code{0}) or at least one moose was observed
+#'     (va ue \code{1}).
+#'   \item geometry: \code{POINT} geometry representing coordinates in an Alaska
+#'     Albers projection (EPSG: 3338).
 #' }
 #' @source
 #' Alaska Department of Fish and Game, Division of Wildlife Conservation has released
@@ -138,11 +138,11 @@
 #' @description Locations at which to predict moose counts and presence in Alaska, USA.
 #'
 #' @format An \code{sf} object with 100 rows and 3 columns.
-#' \describe{
-#'   \item{elev: }{The elevation.}
-#'   \item{strat: }{A factor representing strata (used for sampling). Can take values \code{L} and \code{M}.}
-#'   \item{geometry: }{\code{POINT} geometry representing coordinates in an Alaska
-#'     Albers projection (EPSG: 3338).}
+#' \itemize{
+#'   \item elev: The elevation.
+#'   \item strat: A factor representing strata (used for sampling). Can take values \code{L} and \code{M}.
+#'   \item geometry: \code{POINT} geometry representing coordinates in an Alaska
+#'     Albers projection (EPSG: 3338).
 #' }
 #' @source
 #' Alaska Department of Fish and Game, Division of Wildlife Conservation has released

R/dispersion_initial.R

@@ -17,13 +17,13 @@
 #'   The variance function of an individual \eqn{y} (given \eqn{\mu})
 #'   for each generalized linear model family is given below:
 #'   \itemize{
-#'     \item{family: }{\eqn{Var(y)}}
-#'     \item{poisson: }{\eqn{\mu \phi}}
-#'     \item{nbinomial: }{\eqn{\mu + \mu^2 / \phi}}
-#'     \item{binomial: }{\eqn{n \mu (1 - \mu) \phi}}
-#'     \item{beta: }{\eqn{\mu (1 - \mu) / (1 + \phi)}}
-#'     \item{Gamma: }{\eqn{\mu^2 / \phi}}
-#'     \item{inverse.gaussian: }{\eqn{\mu^2 / \phi}}
+#'     \item family: \eqn{Var(y)}
+#'     \item poisson: \eqn{\mu \phi}
+#'     \item nbinomial: \eqn{\mu + \mu^2 / \phi}
+#'     \item binomial: \eqn{n \mu (1 - \mu) \phi}
+#'     \item beta: \eqn{\mu (1 - \mu) / (1 + \phi)}
+#'     \item Gamma: \eqn{\mu^2 / \phi}
+#'     \item inverse.gaussian: \eqn{\mu^2 / \phi}
 #'   }
 #'   The parameter \eqn{\phi} is a dispersion parameter that influences \eqn{Var(y)}.
 #'   For the \code{poisson} and \code{binomial} families, \eqn{\phi} is always

R/dispersion_params.R

@@ -11,13 +11,13 @@
 #' @details The variance function of an individual \eqn{y} (given \eqn{\mu})
 #'   for each generalized linear model family is given below:
 #'   \itemize{
-#'     \item{family: }{\eqn{Var(y)}}
-#'     \item{poisson: }{\eqn{\mu \phi}}
-#'     \item{nbinomial: }{\eqn{\mu + \mu^2 / \phi}}
-#'     \item{binomial: }{\eqn{n \mu (1 - \mu) \phi}}
-#'     \item{beta: }{\eqn{\mu (1 - \mu) / (1 + \phi)}}
-#'     \item{Gamma: }{\eqn{\mu^2 / \phi}}
-#'     \item{inverse.gaussian: }{\eqn{\mu^2 / \phi}}
+#'     \item family: \eqn{Var(y)}
+#'     \item poisson: \eqn{\mu \phi}
+#'     \item nbinomial: \eqn{\mu + \mu^2 / \phi}
+#'     \item binomial: \eqn{n \mu (1 - \mu) \phi}
+#'     \item beta: \eqn{\mu (1 - \mu) / (1 + \phi)}
+#'     \item Gamma: \eqn{\mu^2 / \phi}
+#'     \item inverse.gaussian: \eqn{\mu^2 / \phi}
 #'   }
 #'   The parameter \eqn{\phi} is a dispersion parameter that influences \eqn{Var(y)}.
 #'   For the \code{poisson} and \code{binomial} families, \eqn{\phi} is always

R/fitted.R

@@ -6,7 +6,8 @@
 #' @param object A fitted model object from [splm()], [spautor()], [spglm()], or [spgautor()].
 #' @param type \code{"response"} for fitted values of the response, \code{"spcov"}
 #'   for fitted values of the spatial random errors, or \code{"randcov"} for
-#'   fitted values of the random effects. The default is \code{"response"}.
+#'   fitted values of the random effects. If from \code{spglm()} or \code{spgautor()},
+#'   \code{"link"} for fitted values on the link scale. The default is \code{"response"}.
 #' @param ... Other arguments. Not used (needed for generic consistency).
 #'
 #' @details When \code{type} is \code{"response"}, the fitted values

R/get_local_list.R

@@ -45,14 +45,14 @@ get_local_list_estimation <- function(local, data, xcoord, ycoord, n, partition_
       if ("groups" %in% names_local) {
         local$size <- ceiling(n / local$groups)
       } else {
-        local$size <- 50
+        local$size <- 100
         local$groups <- ceiling(n / local$size)
       }
     } else {
       local$groups <- ceiling(n / local$size)
     }
     if (!"method" %in% names_local) {
-      local$method <- "random"
+      local$method <- "kmeans"
     }
     local$index <- get_local_estimation_index(local, data, xcoord, ycoord, n)
   }
@@ -105,7 +105,7 @@ get_local_list_prediction <- function(local) {
 
   if (is.logical(local)) {
     if (local) {
-      local <- list(method = "covariance", size = 50)
+      local <- list(method = "covariance", size = 100)
     } else {
       local <- list(method = "all")
     }
@@ -127,7 +127,7 @@ get_local_list_prediction <- function(local) {
   }
 
   if (local$method %in% c("distance", "covariance") && !"size" %in% names_local) {
-    local$size <- 50
+    local$size <- 100
   }
 
   if (!"parallel" %in% names_local) {