The popular specr package in R is good. However, specr is not super flexible—it does not support fixed-effect regressions, and mixed-effect regressions take some work.

I wanted to create a package that would build on specr while being more user-friendly. To that end, I’m working on this set of functions, which I’ve given the bad title of specster:

Code

The current code supports:

Four common models: OLS Linear Regression, Logistic Regression, Fixed-Effect Regression, and Linear Mixed-Effect Regression (i.e., lm, glm, lfe::felm, and lmerTest::lmer in R)
Setup of models
Multicore processing to speed up computation
Graphing with ggplot2

Citations:

specr package, which my repository borrows some from: https://github.com/masurp/specr
Specification Curve Analysis: Simonsohn, U., Simmons, J.P. & Nelson, L.D. (2020). Specification curve analysis. Nature Human Behaviour, 4, 1208–1214. https://doi.org/10.1038/s41562-020-0912-z

Lasso Regression Example

Just an example of Lasso Regression with a short explanation.

.html

Mediation Model Plotting (Work-in-Progress)

This function will draw you a mediation model in ggplot in R. It ain’t awesome, but mediation is kind of silly anyways.
I plan to keep adding to it, as it can only do super simple mediation now.

Code