• Graduate Program
    • Why study Business Data Science?
    • Program outline
    • Courses
    • Course registration
    • Admissions
    • Facilities
      • Student Offices
      • Location
      • Housing
      • Student Council
  • Research
  • News
  • Events
    • Events Calendar
    • Events archive
    • Summer School
      • Behavioral Decision Making
      • Deep Learning
      • Econometrics and Data Science Methods for Business and Economics and Finance
      • Foundations of Data Analysis and Machine Learning in Python
      • Introduction to Genome-Wide Data Analysis
      • Reinforcement Learning
      • Tinbergen Institute Summer School Program
  • Summer School
  • Alumni
Home | Events Archive | Focused Information Criterion for Propensity Score Matching Estimators
Seminar

Focused Information Criterion for Propensity Score Matching Estimators


  • Series
    Econometrics Seminars and Workshop Series
  • Speakers
    Yoshiyasu Rai (University of Mannheim, Germany)
  • Field
    Econometrics
  • Location
    University of Amsterdam and online (hybrid seminar), room E5.22
    Amsterdam
  • Date and time

    May 13, 2022
    12:30 - 13:30

Abstract
This paper studies the model selection problem for propensity score matching estimators of the average treatment effect (ATE) and the average treatment effect on treated (ATET). I derive the large sample distribution of propensity score matching estimators in a local asymptotic framework and characterize the asymptotic bias with respect to the first stage propensity score model choice. I show that the propensity score model choice induces a nontrivial asymptotic bias-variance trade-off for the ATET estimator. I also show that the largest propensity score model achieves the smallest asymptotic mean squared error for the ATE estimator. I then propose a focused information criterion for the propensity score matching estimator of the ATET that aims to minimize the estimated mean squared error. A simulation study indicates that the proposed method generally achieves a smaller mean squared error than other methods.