Рет қаралды 2,721
10 November 2022. "How to interpret and report estimates from (almost) any `R` model? A post-estimation workflow with `marginaleffects` and `modelsummary`." Vincent Arel-Bundock, Associate Professor, Université de Montréal. CANSSI Ontario Statistical Software Conference.
Abstract: The `R` statistical computing ecosystem allows users to fit a wide array of statistical models. Unfortunately, different modeling packages produce inconsistent outputs and objects. This makes it painful and time consuming for users to process, interpret, and report the results of different models. This talk introduces a consistent workflow to interpret and report the results of (almost) any estimation procedure in `R`. This workflow builds on two open source packages. `marginaleffects` includes a suite of post-estimation functions to compute predictions, contrasts, slopes, marginal means, and hypothesis tests. `modelsummary` allows users to draw tables and plots with summary statistics, model estimates, goodness-of-fit statistics, and more. The `marginaleffects` and `modelsummary` packages support over 70 classes of models. They are robust, flexible, and designed to work well with each other and with other packages.