This book aims to fill the gap between panel data econometrics textbooks, and the latest development on 'big data', especially large-dimensional panel data econometrics. It introduces important research questions in large panels, including testing for cross-sectional dependence, estimation of factor-augmented panel data models, structural breaks in panels and group patterns in panels. To tackle these high dimensional issues, some techniques used in Machine Learning approaches are also illustrated. Moreover, the Monte Carlo experiments, and empirical examples are also utilised to show how to implement these new inference methods. Large-Dimensional Panel Data Econometrics: Testing, Estimation and Structural Changes also introduces new research questions and results in recent literature in this field.<b>Contents:</b> <ul><li>Preface</li><li>About the Authors</li><li>Introduction</li><li>Tests for Cross-Sectional Dependence in Fixed Effects Panel Data Models</li><li>Factor Augmented Panel Data Regression Models</li><li>Structural Changes in Panel Data Models</li><li>Latent-Grouped Structure in Panel Data Models</li><li>Bibliography</li><li>Index</li></ul><br><b>Readership:</b> Targeted readers include advanced undergraduates and PhD students and researchers in economics, statistics and business subjects. This book can be used as a textbook or reference book in an advanced undergraduate or graduate level econometrics course. Correlated Effects;Factor Model;Iterated Principal Components;Structural Change;Common Break;Grouped Pattern;K-means;LASSO;Endogeneity0<b>Key Features:</b><ul><li>It provides a guidebook to PhD students and junior faculty who are interested in how traditional inference methods using economic data are affected by large dimension setups, for e.g., large panels and abundance of data</li><li>Different from other textbooks, this book also discusses details of research techniques related to high dimensional issues in econometrics, in addition to introducing new research questions and results in recent literature</li></ul>
Оглавление
Chihwa Kao. Large-Dimensional Panel Data Econometrics
Preface
About the Authors
Contents
Chapter 1. Introduction
Chapter 2. Tests for Cross-Sectional Dependence in Fixed Effects Panel Data Models
2.1.LM Tests for Cross-Sectional Dependence
2.2.LMP Test in the Raw Data Case
2.3.A Bias-Corrected LM Test in a Fixed Effects Panel Data Model
2.4.Dynamic Panel Data Models
2.5.Monte Carlo Simulations
2.5.1.Experiment design
2.5.2.Results
2.6.Recent Development
2.7.Technical Details
2.8.Exercises
Chapter 3. Factor-Augmented Panel Data Regression Models. 3.1.Motivation
3.2.CCE Approach
3.3.IPC Approach
3.4.Likelihood Approach
3.5.Other Studies
3.6.An Empirical Example
3.7.Exercises
Chapter 5. Latent-Grouped Structure in Panel Data Models. 5.1.Panel Latent Group Structure Models
5.2.K-means Clustering
5.3.Conclusion
5.4.Exercises
Bibliography
Index
Отрывок из книги
With the availability of Big Data, one may have more information to identify the underlying causality of economic relationship or forecast important macroeconomic variables or indicators. However, when large volume of data is involved, large dimension could be an issue in the statistical inference of traditional regression models. This book is motivated by the recent development in panel data models with large individuals/countries (n) and large amount of observations over time (T). It introduces testing for cross-sectional dependence and structural breaks in large panels. This book also summarizes important advancement in estimating factor-augmented panel data models and group patterns in panels in recent literature.
This book can be considered complementary to popular panel data econometrics textbooks such as Baltagi (2013), Hsiao (2014) and Pesaran (2015). It is designed for high-level graduate courses in econometrics and statistics. It can be used as a reference for researchers. In specific, Chapters 2 and 4 drew heavily from our published works with Badi H. Baltagi. Chapters 3 and 5 summarize important methods from the recent literature. We would like to thank Badi H. Baltagi for his collaborative work that stimulated our interest in writing this book. We would also like to thank Kunpeng Li for sharing his code, which is used to produce empirical results in Chapter 3. Wei Wang and Mengying Yuan are also acknowledged for helping read the drafts and research assistance. We also wish to thank World Scientific Publishing for giving us the opportunity to undertake this work.
