【Metrics Seminar】Optimal Linear Rank Indexes for Latent Variables

  • 研討會日期 : 2023-06-20
  • 時間 : 10:30
  • 主講人 : Professor Ji-Liang Shiu (徐吉良)
  • 地點 : Conference Room B110
  • 演講者簡介 : Professor Ji-Liang Shiu received his PhD in Economics from Johns Hopkins University in 2009. He is currently a Professor at Jinan University. His resesarch interests are Econometrics and Labor Economics.
  • 演講摘要 : A comprehensive understanding of the quality and performance differences among individuals and institutions is of great importance across many economic applications. However, this is inherently difficult because instead of observing true quality and performance measures directly, analysts often have access only to a set of proxy measures. In this paper, we develop a linear index model to rank the values of a latent variable given at least three measurements. Monte Carlo simulations show that our proposed sieve-based optimal linear rank index estimator performs well in finite samples. We apply our method to assess true GDP growth across countries using various measurements of economic activity and we find that satellite-recorded luminosity and Google search volume index can be important complements to official GDP statistics.