Reproducible Econometrics Using R offers an in-depth exploration of key econometric topics combined with practical guidance on implementing these methods using open-source software (R). Authored by Jeffrey S. Racine, the book is published by Oxford University Press and was released on December 24, 2018. It covers five essential areas: linear time series models, robust inference, robust estimation, model uncertainty, and advanced topics.

The book begins with an introduction to random walks, white noise, and non-stationarity in the context of time series analysis, contrasting it with cross-sectional analysis where the focus is on parameters with economic interpretations. It delves into the pitfalls of using standard inferential procedures for time series data and introduces alternative methods that form the basis of linear time series analysis.

For robust inference, Racine discusses bootstrapping and Jackknifing as alternatives to asymptotic theory, providing a range of numerical approaches. In robust estimation, he addresses issues related to outliers in data and methods for handling them effectively. The section on model uncertainty covers two primary approaches: model selection and model averaging.

Throughout the text, there is a strong emphasis on the benefits of using R and other open-source tools to ensure reproducibility. Advanced topics include machine learning methods such as support vector machines for classification and nonparametric kernel regression for dealing with model uncertainty. This book is well-suited for both advanced undergraduate and graduate students and includes assignments, exams, slides, and a solution manual for instructors.

ISBN-13: 978-0190900687

Publisher: Oxford University Press

Publication Date: December 24, 2018