Occasional Divergences

Coding in R and Python with Positron

Positron is a “next-generation data science IDE.” If RStudio is too specific and VS Code is too general, is Positron just right?

Reproducible environments for R and Python

Reproducible environments are a pain, but {renv} makes things easy for R and works well with Python. Add pyenv and some venv, and you should be set for both R and Python.

Coding discrete explanatory variables

bayesian inference

There are a number of ways to code discrete explanatory variables. In this post, we explore the equivalence of using two of the most common ways: dummy coding and index…

Quarto questions

Everyone is talking about Quarto. Maybe you’ve read some blogs or watched a presentation or two, but you still have questions. I’ve attempted to (briefly) provide some…

Comparing choice model parameterizations

hierarchical models

bayesian inference

Since choice is typically manifest as a non-binary discrete outcome and we care about modeling consumer heterogeneity, a hierarchical Bayesian multinomial logit model is our…

Hierarchical models in Stan with a non-centered parameterization

hierarchical models

bayesian inference

We build a hierarchical linear model with a multivariate population model. This added complexity will require us to re-express the model using what is known as a…

Hierarchical models in Stan

hierarchical models

bayesian inference

A gentle introduction to building hierarchical models in Stan via R and using a tidy approach whenever appropriate. This is not an introduction to Bayesian inference or…