Latent Class Regression Tool

Estimate segment-level regression models using the flexmix package

Disclaimer

Results may contain errors. The output from this tool may be inaccurate, incomplete, or incorrect. Users should independently verify all results before relying on them.
No warranty. This tool is provided "as is" without any warranty of any kind.
User assumes all risk. Users assume full and complete responsibility for reviewing, validating, and determining suitability of any output.

Requirements

R - Statistical programming language (Download R)
RStudio - IDE for R (Download RStudio)

Setup Instructions

Install R and RStudio

If you don't have R and RStudio installed, download and install them from the links above. Install R first, then RStudio.

Download and Open the App

Download the app file and open it in RStudio:

In RStudio: File → Open File → Select the downloaded Latent_Class_Regression.R

Run the App

With Latent_Class_Regression.R open in RStudio, click the "Run App" button in the top-right corner of the editor panel. The app will open in a new window or your web browser.

Note: On first run, the app will automatically install the required R packages (shiny and flexmix). This may take a minute.

Tip: To open in fullscreen/external browser, click the "Open in Browser" button in the app window, or use the dropdown next to "Run App" and select "Run External".

Features

CSV Data Upload

Upload your dataset in CSV format

Three Model Types

Linear, Poisson, or Binary Logistic regression

Segment Selection

Specify the number of latent segments (1-10)

Consumer/HH ID Grouping

Cluster observations by Consumer or Household ID

Segment Estimates

View coefficients, std. errors, and p-values for each segment

Model Fit Statistics

AIC, BIC, and segment sizes (prior probabilities)

Troubleshooting

App doesn't open

Make sure you have the Latent_Class_Regression.R file open in RStudio's editor (not just in the file browser), then click "Run App".

Package installation fails

Try running these commands manually in the Console:

install.packages("shiny")
install.packages("flexmix")
                

Model fails to converge

Try reducing the number of segments, or check that your data has enough variation. The flexmix algorithm may need more observations per segment.

"Run App" button not visible

Make sure the file is saved with a .R extension. RStudio only shows the "Run App" button for R files that contain Shiny code.