Software products developed as part of our research are always made available as part of online supplementary materials through journal websites where the research gets published. Additional software products are sometimes released separately with significant added features, substantial performance improvements, and enhanced user friendliness. Such products and related (peer-reviewed) papers released or published separately are listed here.

We do not provide personal software support. Version updates, if any, will be made available through this website or, for official R releases, the Comprensive R Archive Network (CRAN).

1. R package BayesDecon (2026): Implements Bayesian semiparametric univariate and multivariate density deconvolution for continuous and zero-inflated data. Provides a general-purpose solution for recovering latent distributions under unknown, conditionally heteroscedastic error structures, leveraging replicated (potentially zero-inflated) proxies per sample unit. Also supports fitting simpler sub-models, including homoscedastic noise and/or pre-specified parametric error distributions (e.g., Gaussian or Laplace). Uses National Health and Nutrition Examination Survey (NHANES) dietary data as a testbed for validation and illustration.
   Main developers: Blake Moya and Mainak Manna [link to the package on CRAN]
   


2. R package heiscore (2024): Implements three different scoring methods recommended by the Dietary Guidelines of America (DGA) for analyzing the Healthy Eating Index (HEI) using the National Health and Nutrition Examination Survey (NHANES) data, facilitating the visualization of component scores and their comparisons across different population subgroups and individuals.
   Main developers: Berkeley Ho and Vijetha Ramdas [link to the package on CRAN]
   Paper on the package: Ho, B. N., Ramdas, V. and Sarkar, A. (2024+) HEI analysis of NHANES dietary data: Exploring the diet quality of Americans with R package heiscore. To appear in the Journal of Data Science. [link]


3. R package BMRMM (2022): Implements flexible Bayesian Markov (renewal) mixed models to analyze categorical sequences (and associated state duration times or inter-state intervals).
   Main developer: Yutong Wu
   [link to methodology paper 1] [link to methodology paper 2] [link to the package on CRAN]
   Paper on the package: Wu, Y. and Sarkar, A. (2024). BMRMM: An R package for Bayesian Markov (renewal) mixed models. R Journal, 16, 192-211. [link]


4. R package lddmm (2022): Implements a flexible Bayesian longitudinal drift-diffusion mixed model for category learning to ana- lyze response accuracies and associated (censored) response times.
   Main developer: Giorgio Paulon
   [link to the methodology paper] [link to the package on CRAN]