Software

Statistical Software

A list of Stata commands and R packages I have written. Click on the package name below to read more about each (including links to download):

Stata

• balanceplot: plots of standardized imbalance statistics
• cleanplots: Stata graphics scheme for professional and accessible graphs 
• desctable: professionally formatted descriptive statistics tables
• irt_coef: y*-standardized coefficient estimates for item response theory models
• irt_me: marginal effect estimates for item response theory models
• lca_entropy: entropy fit statistic calculation for latent class analysis models
• mecompare: comparing marginal effects across models
• meinequality: marginal effects inequality statistics
• metest: custom tests of marginal effects and other estimates
• sgmediation2: Stata command for Sobel-Goodman tests of statistical mediation in linear regression 
• suest2: seemingly-unrelated estimation for most regression models in Stata
• totalme: total marginal effects estimates
• usetdm: command to download datasets off web for class and workshop examples

R

• cleanplots: cleanplots color scales and theme for ggplot2
• suest: seemingly unrelated estimation for comparing predictions and marginal effects across models
• meinequality: marginal effects inequality statistics in R, with marginaleffects
• totalme: total marginal effects in R, with marginaleffects

Misc. Example Stata Do-Files

Below are a host of miscellaneous example Stata do-files I have collected over the years. These are examples I have created for things I do repeatedly in my own work, and/or examples I have created for common questions I receive from people looking for help.

Note: For most topics, you will find more extensive and more polished examples associated with a class I teach.

Disclaimer: These templates worked for me for the specific situation I faced. Consider these as starting points for your work –not guaranteed paths to success.

• Various ways to fit a fixed effects linear model: demeaning, hybrid, between/within, correlated random effects
• Using marginal effects and mimrgns with imputed data: main effects, interaction effects, mediation, and other cross-model comparisons
• Impute missing data with multiple hot deck, including a stratification variable that needs to be imputed
• Examine interaction effects among predictors of latent class membership
• Data cleaning template (empty, starting file)
• Data analysis template (empty, starting file)