Package to infer population level eccentricity distributions using hierarchical MCMC.
First, (make sure you have orbitize! installed) [https://orbitize.readthedocs.io/en/latest/installation.html].
Next, from the commmand line:
$ git clone https://github.com/vighnesh-nagpal/ePop.git
$ cd ePop
$ pip install -e . --upgrade
import ePop.simulate
import ePop.hier_sim
# simulate a forward modelled sample of 10 imaged companion eccentricity posteriors
# drawn from the RV exoplanet distribution from Kipping (2010)
a, b = 0.87, 3.03
# this step simulates realistic eccentricity posteriors for a set of systems with
# eccentricities drawn from the (0.87, 3.03) Beta Distribution. (Time-Intensive)
ecc_posteriors=ePop.simulate.simulate_sample((a,b))
# create Likelihood object and choose hyperprior
like=ePop.hier_sim.Pop_Likelihood(posteriors=ecc_posteriors,prior='Gaussian')
# NOTE: you can also load in samples from already saved eccentricity posteriors and
# initialise a likelihod object as above. In this case, posteriors must be a list of
# 1D arrays, where the arrays are the 1D eccentricity posteriors for each system in the sapmle.
# sample the hyperparameters using MCMC using 1000 steps.
beta_samples=like.sample(1000,burn_steps=500,nwalkers=30)
