Premium
Random forests and selected samples
Author(s) -
Cook Jonathan A.,
Siddiqui Saad
Publication year - 2020
Publication title -
bulletin of economic research
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.227
H-Index - 29
eISSN - 1467-8586
pISSN - 0307-3378
DOI - 10.1111/boer.12222
Subject(s) - selection (genetic algorithm) , monte carlo method , random forest , parametric statistics , nonlinear system , outcome (game theory) , computer science , econometrics , distribution (mathematics) , mathematical optimization , mathematics , random variable , statistics , artificial intelligence , mathematical economics , quantum mechanics , mathematical analysis , physics
This paper presents a procedure for recovering causal coefficients from selected samples that uses random forests, a popular machine‐learning algorithm. This proposed method makes few assumptions regarding the selection equation and the distribution of the error terms. Our Monte Carlo results indicate that our method performs well, even when the selection and outcome equations contain the same variables, as long as the selection equation is nonlinear. The method can also be used when there are many variables in the selection equation. We also compare the results of our procedure with other parametric and semiparametric methods using real data.