Andrew Heiss is a researcher and educator focused on data visualization, causal inference, and applied statistics using R and Bayesian methods. He writes extensively about reproducible research, GIS, and analytical workflows, and teaches data science and social science methods.
55 articles from this blog
A guide to using Bayesian beta and zero-inflated beta regression models in R to correctly analyze proportion data.
Explains the three rules of do-calculus in plain language and manually derives the backdoor adjustment formula for causal inference.
A guide to converting TikZ diagrams to SVG with embedded fonts using knitr and R Markdown for both PDF and HTML outputs.
A tutorial on using R to diagnose biases in two-way fixed effects (TWFE) regression models when treatment timing varies, based on Pamela Jakiela's paper.
A tutorial using Euler/Venn diagrams to visualize and explain the R² statistic and variance in regression models.
A technical guide on using marginal structural models with GEE and multilevel models in R to handle confounding in panel data.
A technical tutorial on using R and inverse probability weighting to handle time-series panel data for causal inference with marginal structural models.
A tutorial on using R to calculate inverse probability weights for causal inference with both binary and continuous treatment variables.
Explains methods like regression and inverse probability weighting to close confounding backdoors in DAGs for causal inference in observational data.
A tutorial on using a Makefile to automatically zip subdirectories, handling dependencies and excluding hidden files for tech projects.
A tutorial on building an interactive data dashboard using R, flexdashboard, and Shiny with data from Google Sheets.
A guide to creating a macOS service that converts Markdown to rich text with syntax highlighting for use in any app, like email clients.
Using R and the yacas library to solve a microeconomics optimization problem: finding the optimal consumption of pizza and yogurt given a budget and utility function.
A guide to six statistical methods (frequentist and Bayesian) for comparing group means, with R and Stan code examples.
A tutorial on using the infer package in R for hypothesis testing through simulation, following a modern statistical approach.
