Review of Introductory Econometrics: Using Monte Carlo Simulation with Microsoft Excel (Hardcover)

When I was a new graduate student I ended up buying several different econometrics texts. No one text had the best explanation for each topic. The problem remained that for many topics I never did find a book which translated the formal mathematical presentation into a practical worked out example, so that I could understand the procedure and how to implement it.

This book and accompanying CD-ROM does that and much more.

Every topic includes guided Microsoft Excel spreadsheets and add-ins which illustrate the topic being addressed. The text clearly explains not only the HOW, but the WHY. The economics and the econometrics are presented with such clarity and unity; bridging the two in a way that none of the other texts do.

In Barreto and Howlands book/CD package you interact with the data and the graphs (they include a superior add-in for creating histograms), and run Monte Carlo simulations to see the behavior of the estimators in repeated sampling. These are "live" spreadsheets that invite you to experiment. For example; there is an Excel workbook which illustrates the correlation coefficient. Rather than a dry recitation of formula and proof, you can interact with the spreadsheet and see exactly how the same coefficient can apply to data having very different patterns. It is one thing to see an illustration, and quite another to actually be the one creating the diagram, simply by running the macros and changing parameters. This "hands on" approach is so vital to actually getting an understanding of the material. I have only a basic understanding of Excel, and have had no difficulties in using the workbooks.

While the limits of Excel are pointed out by the authors, it is important to note the reason for using Excel. It is widely understood and available; there is no learning curve. By using Excel there is no software barrier between the student and understanding the principles of econometric modelling. In less than 1/2 hour I took the data and example of a Probit model using Maximum Likelihood estimation from a course web site from across the country and replicated the results using the add-in provided.Most of that time was used to extract the data from a .pdf file and get it formatted properly for Excel. Once I had the data in Excel, it took less than 2 minutes to run the Probit estimation (my first time using that add-in!) By the way, the results using the authors add-in for solving Probit models with ML estimation were the same as the results from GAUSS code to do the same. The add-in had a distinct advantage though in that a choice for Probit or Logit model estimation using either Non Linear Least Squares or the Maximum Likelihood estimation was just a radio button away! This text can complement any course, regardless of the software used.

Again, the beauty of the book is that you are not just left with Greek formulas that leave you wondering how to do the computations, and you are not left with computer output leaving you to wonder how to interpret that output. The text explains the meaning so powerfully that you are not only armed with an understanding which is useful for success in your course work, but also for applying the quantitative tools in real world analysis and applications.The text is like going to see your favorite professor who is sitting there with you one on one, giving you insight which only comes from experience.

I've been through courses that use Greene, and Judge, as well as introductory texts. This text stands alone in making use of the computing power we have at our disposal today, not to produce more computer printouts, but rather to increase our understanding--providing the sound reasoning for applying that power.

I should add that even after two years of statistics and econometrics I learned quite a lot from the statistics review chapters. Don't be misled by the "Introductory" title. I had learned and executed Artificial Neural Network models in graduate courses, but still learned a lot from the section on correlation in this book. For undergrad students this book will put you on the right path. For grad students it will correct blind spots and misconceptions.

I highly recommend this book/CD package to any econometrics student and to practicing analysts that use regression analysis. The authors have created a product that I wish I had when I was in school, but am glad I found now for applying in my career.

Detailed info on contents as well as the Excel files and add-ins are available from the authors' web site which I found prior to ordering from Amazon. Once I tried the workbooks, I knew I wanted the book. It's 800 pages of solid information and inspired teaching.

http://www.wabash.edu/econometrics/index.htm



Click Here to see more reviews about: Introductory Econometrics: Using Monte Carlo Simulation with Microsoft Excel (Hardcover)