167 – 167 - Data Mining and Econometrics
Identification and Estimation of Demand in Large Concentrated Markets
Saman Banafti
University of California, Riverside
Gabaix and Koijen (2020) introduces the Granular Instrumental Variables (GIV) methodology, which takes advantage of panel data to construct instruments to estimate structural time series regression models that involve endogenous regressors. The GIVs are constructed based on panel data models with factor structures, where the idiosyncratic error terms may have extraordinarily useful information. In this paper, we extend their GIV methodology by developing the GIV identification procedure to a large $N$ and large $T$ framework (current identification is for fixed $N$ and large $T$) by establishing and restricting the asymptotic behavior of the Herfindahl index for large $N$ markets as a function of the tail index of the size distribution of the cross-sectional units.