Ensemble Methods for Causal Effects in Panel Data Settings

13 Pages Posted: 26 Mar 2019

See all articles by Susan Athey

Susan Athey

Stanford Graduate School of Business

Mohsen Bayati

Stanford Graduate School of Business

Guido W. Imbens

Stanford Graduate School of Business

Zhaonan Qu

Stanford University

Date Written: March 2019

Abstract

In many prediction problems researchers have found that combinations of prediction methods (“ensembles”) perform better than individual methods. A simple example is random forests, which combines predictions from many regression trees.A striking, and substantially more complex, example is the Netflix Prize competition where the winning entry combined predictions using a wide variety of conceptually very different models. In macro-economic forecasting researchers have often found that averaging predictions from different models leads to more accurate forecasts. In this paper we apply these ideas to synthetic control type problems in panel data setting. In this setting a number of conceptually quite different methods have been developed, with some assuming correlations between units that are stable over time, others assuming stable time series patterns common to all units, and others using factor models. With data on state level GDP for 270 quarters, we focus on three basic approaches to predicting missing values, one from each of these strands of the literature. Rather than try to test the different models against each other and find a true model, we focus on combining predictions based on each of the separate models using ensemble methods. For the ensemble predictor we focus on a weighted average of the three individual methods, with non-negative weights determined through out-of-sample cross-validation.

Institutional subscribers to the NBER working paper series, and residents of developing countries may download this paper without additional charge at www.nber.org.

Suggested Citation

Carleton Athey, Susan and Bayati, Mohsen and Imbens, Guido W. and Qu, Zhaonan, Ensemble Methods for Causal Effects in Panel Data Settings (March 2019). NBER Working Paper No. w25675, Available at SSRN: https://ssrn.com/abstract=3359432

Susan Carleton Athey (Contact Author)

Stanford Graduate School of Business ( email )

655 Knight Way
Stanford, CA 94305-5015
United States

Mohsen Bayati

Stanford Graduate School of Business ( email )

655 Knight Way
Stanford, CA 94305-5015
United States

HOME PAGE: http://web.stanford.edu/~bayati/

Guido W. Imbens

Stanford Graduate School of Business ( email )

655 Knight Way
Stanford, CA 94305-5015
United States

Zhaonan Qu

Stanford University

Here is the Coronavirus
related research on SSRN

Paper statistics

Downloads
14
Abstract Views
238
PlumX Metrics