High-Dimensional Econometrics and Identification (ePub)

In many applications of econometrics and economics, a large proportion of the questions of interest are identification. An economist may be interested in uncovering the true signal when the data could be very noisy, such as time-series spurious regression and weak instruments problems, to name a few. In this book, High-Dimensional Econometrics and Identification, we illustrate the true signal and, hence, identification can be recovered even with noisy data in high-dimensional data, e.g., large panels. High-dimensional data in econometrics is the rule rather than the exception. One of the tools to analyze large, high-dimensional data is the panel data model.

High-Dimensional Econometrics and Identification grew out of research work on the identification and high-dimensional econometrics that we have collaborated on over the years, and it aims to provide an up-todate presentation of the issues of identification and high-dimensional econometrics, as well as insights into the use of these results in empirical studies. This book is designed for high-level graduate courses in econometrics and statistics, as well as used as a reference for researchers.

• Preface
• About the Authors
• Panel Data Model with Stationary and Nonstationary Regressors and Error Terms
• Panel Time Trend Model with Stationary and Nonstationary Error Terms
• Estimation of Change Points in Stationary and Nonstationary Regressors and Error Term
• Weak Instruments in Panel Data Models
• Incidental Parameters Problem in Panel Data Models
• Bibliography
• Index

• This book focuses on panel data models with both large cross-sectional dimension, n, and time-series dimension T
• Existing panel data textbooks, such as Baltagi (2013), Hsiao (2014) and Pesaran (2015), usually study panel data models with a large dimension n but a fixed dimension T. Different from them, we show in this book that identification can be restored in a panel data with large dimensions n and T