Skip to contents

shinymrp: Applying Multilevel Regression and Poststratification in R

shinymrp allows users to apply Multilevel Regression and Poststratification (MRP) methods to a variety of datasets, from electronic health records to sample survey data, through an end-to-end Bayesian data analysis workflow. Whether you’re a researcher, analyst, or data engineer, shinymrp provides robust tools for data cleaning, exploratory analysis, flexible model building, and insightful result visualization.

  • Data preparation: Clean, preprocess and display the input data.
  • Descriptive statistics: Visualize key summary statistics.
  • Model building: Specify and fit models with various predictors as fixed or varying effects. Guide your model selection with detailed model diagnostics and comparison metrics.
  • Result visualization: Generate graphs to convey population-level and subgroup estimates, facilitating interpretation and communication of your findings.

Getting Started

You can use shinymrp in two flexible ways:

Shiny App

The graphical user interface (GUI), built with the Shiny framework, is designed for newcomers and those looking for an interactive, code-free analysis experience.

Launch the app locally in R with:

shinymrp::run_app()

Try the Demo

Explore the Shiny app without installation via our online demo.

Need a walk-through? Watch our step-by-step video tutorial.

Object-Oriented Programming Interface

Leverage the full flexibility of the exported R6 classes for a programmatic workflow, ideal for advanced users and those integrating MRP into larger R projects.

Import shinymrp in scripts or R Markdown documents just like any other R package:

Installation

To get started, install the latest development version of shinymrp from GitHub using remotes:

# If 'remotes' is not installed:
install.packages("remotes") 
remotes::install_github("mrp-interface/shinymrp")

The package installation does not automatically install all prerequisites. Specifically, shinymrp uses CmdStanR as the bridge to run Stan, a state-of-the-art platform for Bayesian modeling. Stan requires a modern C++ toolchain (compiler and GNU Make build utility).

Learn More

For detailed guidance, check our introductory vignette: Getting started with shinymrp.

This product uses the Census Bureau Data API but is not endorsed or certified by the Census Bureau.