Comment from the Stata technical group
Introduction to Time Series Using Stata, Revised Edition, by Sean Becketti, is a practical guide to working with timeseries data using Stata. In this book, Becketti introduces timeseries techniques—from simple to complex—and explains how to implement them using Stata. The many worked examples, concise explanations that focus on intuition, and useful tips based on the author’s experience make the book insightful for students, academic researchers, and practitioners in industry and government.
Becketti is a financial industry veteran with decades of experience in academics, government, and private industry. He was also a developer of Stata in its infancy and has been a regular Stata user since its inception. He wrote many of the first timeseries commands in Stata. With his abundant knowledge of Stata and extensive experience with realworld timeseries applications, Becketti provides readers with unique insights and motivation throughout the book.
For those new to Stata, the book begins with a mild yet fastpaced introduction to Stata, highlighting all the features you need to know to get started using Stata for timeseries analysis. Before diving into analysis of time series, Becketti includes a quick refresher on statistical foundations such as regression and hypothesis testing.
The discussion of timeseries analysis begins with techniques for smoothing time series. As the movingaverage and Holt–Winters techniques are introduced, Becketti explains the concepts of trends, cyclicality, and seasonality and shows how they can be extracted from a series. The book then illustrates how to use these methods for forecasting. Although these techniques are sometimes neglected in other timeseries books, they are easy to implement, can be applied quickly, often produce forecasts just as good as more complicated techniques, and, as Becketti emphasizes, have the distinct advantage of being easily explained to colleagues and policy makers without backgrounds in statistics.
Next, the book focuses on singleequation timeseries models. Becketti discusses regression analysis in the presence of autocorrelated disturbances as well as the ARIMA model and Box–Jenkins methodology. An entire chapter is devoted to applying these techniques to develop an ARIMAbased model of U.S. GDP; this will appeal to practitioners, in particular, because it goes step by step through a realworld example: here is my series, now how do I fit an ARIMA model to it? The discussion of singleequation models concludes with a selfcontained summary of ARCH/GARCH modeling.
In the final portion of the book, Becketti discusses multipleequation models. He introduces VAR models and uses a simple model of the U.S. economy to illustrate all key concepts, including model specification, Granger causality, impulse–response analyses, and forecasting. Attention then turns to nonstationary timeseries. Becketti masterfully navigates the reader through the oftenconfusing task of specifying a VEC model, using an example based on construction wages in Washington, DC, and surrounding states.
Introduction to Time Series Using Stata, Revised Edition, by Sean Becketti, is a firstrate, examplebased guide to timeseries analysis and forecasting using Stata. This is a musthave resource for researchers and students learning to analyze timeseries data and for anyone wanting to implement timeseries methods in Stata.
Comments from readers
I have been producing forecasts and using Stata for many years, but I still learned something new on almost every page. But don't be intimidated, the book has stepbystep instructions and easytofollow examples for those who are new to Stata or time series analysis or both.
Patrick J. Flaherty
Connecticut Department of Labor
