autoMrP improves the prediction performance of multilevel regression with post-stratification (MrP) by combining a number of machine learning methods through ensemble Bayesian model averaging (EBMA). The full discussion can be found here: Broniecki, Leemann, and Wüest. 2020. “Improving Multilevel Regression with Post-Stratification Through Machine Learning (autoMrP),” forthcoming in the Journal of Politics.
To install autoMrP from GitHub run:
Please refer to the vignette for a detailed introduction to autoMrP. Access the vignette via: