Posts

Lost in Transformation: The Horror and Wonder of Logit

Perhaps no other subject in applied statistics and machine learning has caused people as much trouble as the humble logit model. Most people who learn either subject start off with some form of linear regression, and having grasped it, are then told that if they have a variable with only two values, none of the rules apply.

New Survey Research on Tunisia and the IMF Deal

Tunisia’s dictator, President Kais Saied, is pushing the country towards full-scale economic collapse by refusing to negotiate with the International Monetary Fund about a potential rescue deal. Neither the government nor the increasingly repressed opposition wants to take responsibility for the wrenching changes proposed by the IMF despite the offered loan package of \$1.

What Do Tunisians Really Think About President Kais Saied?

Robustness check: time of completion Survey Design Sensitivity Estimates Alternative Estimation: RRreg Alternative Estimator: rr BRMS Model Adjustment with MRP Gender Age Region Simulation Comparison: Univariate Vs.

Fixing Fractional Logit?

This post focuses on one of the more curious models in contemporary statistics, a specification for proportions that is either called fractional logit or quasi-Binomial. An earlier version of this blog post had a much more negative take on the fractional logit specification.

What's Logs Got to Do With It?

Twitter (or what’s left of it) was recently ablaze with a discussion of two smart working papers, one by Jiafeng Chen and Jonathan Roth and the other by John Mullahy and Edward Norton.

Simulating Turnout in Tunisia's Constitutional Referendum

I am writing this post in response to questions about estimating turnout for Tunisia’s constitutional referendum today. Turnout is an important aspect to this referendum because high turnout would signal higher legitimacy for President Kais Saied’s dramatic changes to the Tunisia’s democracy.

What To Do (And Not to Do) with Modeling Proportions/Fractional Outcomes

Interested in more social science on contemporary issues? Check out my just-released book with Cambridge University Press and use discount code KUBINEC23 to get 20% off. Introduction Limited dependent variables, or continuous variables with lower and upper bounds, are quite common in the social sciences but do not fit easily with existing statistical models.

Energy Nationalism To the Rescue

The Russian invasion of the Ukraine has raised a new specter of energy nationalism. Although only the United Kingdom and the United States have gone as far as outright banning Russian oil and gas, other European economies, including Germany, are seriously discussing weaning off of their main energy supplier.

The Causal Representation of Panel Data: A Comment On Xu (2022)

NB: An earlier version of this post critiqued Victor Chernozhukov’s approach to directed a-cyclic graphs and fixed effects, but made some critical errors in interpreting his approach. These errors were entirely mine, and I apologize to Victor for doing so.