Customer Reviews

Introductory Econometrics: Using Monte Carlo Simulation with Microsoft Excel

5 star:		(7)
4 star:		(1)
3 star:		(0)
2 star:		(1)
1 star:		(1)

 

 

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

I am only half finished with this book, but since there is only one other review, I want to get my thoughts up NOW. I may add to them when I have finished.

My wife is an econ major at a small school with very few econ majors. Econometrics is not offered as a course. Although as a practical businessman with a preference for Austrian school economic theory I have a healthy scepticism about quantitative macroeconomic (especially) formulas, I have told my wife that she can not be a part of today's theoretical discussions without a basic understanding of econometrics. I promised to help her self-study this topic, and have reviewed a number of supposedly "introductory" texts (to remain nameless, but they are standards)that have lost me within 50 pages. Neither my wife nor I have calculus or matrix algebra. However, even those texts that say they do not rely on such math knowledge are still confusing. Until now.

Barreto's text is a wonder. The other review gives solid examples of why this is. Let me just say that you will be able to see econometric principles in action. The explanations are incredibly clear, and the work on the beefed up excel spreadsheets effectively demonstrates those explanations. I know this will be difficult to believe, but the text is actually fun to read. My wife and I both have college algebra, business statistics, and basic excel. That's all you need to use this book.

Every university should adopt this book as the intro econometrics text. It provides an approach to learning the topic that is accessible to any intelligent econ student. Those going on to PhD work could supplement with calculus, matrix algebra, and one of the other so-called intro texts. Barreto provides a way for normal econ students to understand econometrics, something that all econ students should be required to do. (Even though much of econometrics is nonsense, knowledge of its applications and mis-applications is still the ticket to being taken seriously in economic debate.)

I only wish I could give this book more than 5 stars. It is a stunning achievement.

Excellent book for everybody who would like to understand statistics and econometrics. It concentrates on regression and Monte Carlo simulation. I see the biggest value of the book in the way the statistical concepts are introduced. The authors build the intuition of the reader using meaningful examples, explaining thoroughly all the concepts and using Monte Carlo simulations to visualize them. Simulations are used as a tool for estimating parameters, but also support the reader's intuition and visualize the stochastic variables, their distributions, the role of chance in statistical inference, hypothesis testing etc. The examples in the book are not just illustration of the text, they are meant to open reader's eyes to some unexpected / counterintuitive features, that are difficult to capture just by studying the theory.

For the numerically inclined, this is one of the best introductory texts I have ever digested. This book is an essential tool for all who practice the art of analysis and are uninitiated with the fundamentals of econometrics.

A very good text for learning about statistical techniques used in econometrics. Very detailed instructions, but easy to follow. I highly recommend the textbook.

As an academic with little background in economics who constantly struggled with advanced techniques, over the years I have tried several econometrics textbooks. Every book has its strengths, but of all books I have been exposed to, I thought this one had the clearest exposition. Concepts are explained well, and the Monte Carlo approach helps tremendously in grasping the underpinning concepts. Things I liked about this book in general:
* It clearly distinguishes between descriptive and inferential statistics - between describing and summarizing data, and drawing inferences from that data.
* Use of Excel sheets to try things out and "see" many things for yourself (for example, see how the least squares achieves best fit to the data).
* It uses real data on economic indicators, such as data on labor markets.
* It minimizes use of matrix algebra or calculus. Mathematical proofs etc. are pushed into appendices at the end of chapters, which can be skipped without loss of continuity.
Things I did not like:
* You have to wait until Chapter 10 in order to get to statistical inference. The first 200 pages is all about summarizing and interpreting patterns in data.
* The book requires you to constantly switch between Excel sheets and the book itself. As a result, every chapter takes longer to finish. Sometimes you have to work on more than one worksheet within a chapter, which adds to this difficulty.
* Excel is not a statistics program and can be cumbersome. Moreover, if the student is to take more advanced courses, Excel will not be enough for him/her.

I think overall the pros far outweigh the cons. I would suggest this book to students who want to grasp the basics well before moving on to more advanced courses; to graduate students outside economics who want to get a good understanding of ordinary least squares (OLS) regression and its assumptions, and to practitioners who want to have a working knowledge of OLS and its limitations. The book would be too introductory for graduate economics students in my opinion. But if you are lost in your `standard' econometrics textbook as an undergraduate, this book may save you. Finally, as with any other textbook, I would pair this book with "A Guide to Econometrics" by Kennedy, which is an excellent supplement.

This book is a perfect introduction to econometrics for practitioners who may be short on foundational theory but know how to use Microsoft Excel. The approach is not traditional, but it is plenty rigorous--a fact that might be disguised by the excellent writing. I recommend this book as a thorough yet accessible resource for self-study.

I give this book such a high score because it really does a great job at presenting the concepts. I know many people who attempt to perform a regression analysis, but have no understanding of what it is they are doing. This book does a great job at presenting what is going on and the effect data generations procresses have on the regression estimate. This book does not provide the detailed mathematics, but who cares. That can be picked from another book on linear modeling. All in all, this book is a good place to start when learning linear modeling.

Where is the application??

The author uses very little, if any, examples of applying econometrics to financial topics. The book is strewn with Larry Bird free-throw stats and SAT scores, height weight distributions and skiing training! I was expecting daily returns on the S&P 500.

I'm giving 2 stars because the CD included has some useful things, although the VB code is very amateurish.

Introductory Econometrics: Using Monte Carlo Simulation with Microsoft Excel
This book is a waste of time and money. It is basically a book on statistics. Verbose but not much teaching. 90% of it I (as well as most people who have studied basic statistics) already know and do not need it. The other 10% I don't know, it does not teach me except by giving me a few button to press.