What's the best way for a forecaster to communicate when their payoff for a binary outcome differs from the decision maker? Guys, it's time for some game theory.
A small note on a statistical result. Some simulations and maximum likelihood estimation with R, Broom and Purrr.
Slides summarizing recent investigations into dynamic model averaging and dynamic logistic regression for binary outcomes. Also, code for making slides.
Back we go into the vasty deep. In this post we look closer at using dynamic model averaging and dynamic logistic regression to forecast recessions.
We go into the vasty deep, dipping our toes ever so slightly into the dark waters of macroeconometric forecasting. Here we use dynamic model averaging to forecast recessions with R.