Post-Selection Inference for Generalized Linear Models with Many Controls

2013/04/15 by Belloni, Alexandre, Chernozhukov, Victor, Wei, Ying · 2 citations
#Econometrics (econ.EM) #FOS: Computer and information sciences #FOS: Economics and business #FOS: Mathematics #Methodology (stat.ME) #Statistics Theory (math.ST)

paper · doi:10.48550/arxiv.1304.3969

Abstract

This paper considers generalized linear models in the presence of many controls. We lay out a general methodology to estimate an effect of interest based on the construction of an instrument that immunize against model selection mistakes and apply it to the case of logistic binary choice model. More specifically we propose new methods for estimating and constructing confidence regions for a regression parameter of primary interest α₀, a parameter in front of the regressor of interest, such as the treatment variable or a policy variable. These methods allow to estimate α₀ at the root-n rate when the total number p of other regressors, called controls, potentially exceed the sample size n using sparsity assumptions. The sparsity assumption means that there is a subset of s

Cited by

• Mixed LR-C(α)-type tests for irregular hypotheses, general criterion functions and misspecified models
• Causal Inference in High-Dimensional Generalized Linear Models with Binary Outcomes

Related

• Post-Selection and Post-Regularization Inference in Linear Models with Many Controls and Instruments
• Inference on Treatment Effects After Selection Amongst High-Dimensional\n Controls
• Variable Selection and Model Averaging in Semiparametric Overdispersed Generalized Linear Models
• Fixed Effects Binary Choice Models with Three or More Periods
• Post-selection inference for linear mixed model parameters using the conditional Akaike information criterion