Introduction

bayes_traj is a software package written in Python that provides routines for performing Bayesian trajectory modeling of longitudinal data. Multiple, longitudinally observed target variables -- continuous, binary, or a combination -- can be modeled simultaneously. Per-trajectory random effects can also be modeled for continuous target variables. This package also provides command-line tools that facilitate spefication of Bayesian priors, enable visualization of trajectory modeling results, and compute summary and model fit statistics.

Installation

In order to install the package, type the folowing in the terminal:

$ pip install bayes_traj

Overview

bayes_traj provides several command-line tools:

• generate_prior -- used to speficy Bayesian priors for use the trajectory modeling
• viz_data_prior_draws -- provides visualization of random draws from the prior
• bayes_traj_main -- performs Bayesian trajectory modeling using a prior file
• viz_model_trajs -- provides visualization of trajectories fit using bayes_traj_main
• sumarize_traj_model -- prints model summary and fit statistics given a model file produce by bayes_traj_main
• assign_trajectory -- writes a data file with appended trajectory assignment information given an input data file and a model file generated by the bayes_traj_main tool

Each of these tools can be run with the -h flag for additional usage information.

For additional documentation, see https://acil-bwh.github.io/bayes_traj/index.html

Tests

To run all unit tests, type the following in the package root directory:

$ pytest

Contribute

Please read our contribution guidelines.