Saturday, January 29, 2011

Book Review: Microsoft Excel 2010: Data Analysis and Business Modeling by Wayne L. Winston

My rating: 4 of 5 stars

When Microsoft Excel ® 2010 came out, I noted that it claimed to finally correct long standing errors in its statistical functions (1). While I wait for McCoullough et al (2) to do their usual reviews, I was encouraged to see that it now passes some basic statistics tests where previous versions consistently failed. For all its faults, spreadsheets are the principle vehicle of communicating quantitative information in business. And to support that, Microsoft Excel has built in many tools for modeling and data analysis. But discussion for many of these techniques in MS Excel books is limited to describing the dialog boxes that are needed to use the functionality. Winston in Microsoft Excel 2010 Data Analysis and Business Modeling provides a needed in depth look at Excel as a data analysis platform in addition to a spreadsheet.

Winston wrote this book as a result of teaching MBA and consulting clients spreadsheet modeling classes. It is broken up into 84 short chapters, each comes with examples (using downloadable spreadsheets and data) and example problems. Each chapter is used to demonstrate a defined concept. For example, financial issues are covered in three chapters, demonstrating net present value, internal rate of return, and the standard bond/loan finance calculations.

The many short chapters allows topics to be covered in some depth. Seven short chapters cover random numbers, followed by five chapters on simulation and some uses of simulation. While the first chapter on a topic may cover the steps to use the tool, the followon chapters all present uses. For example, the Solver is introduced in Chapter 28 with an example that steps through the menu and dialog boxes. Then each of the following seven chapters describe the use of the solver for a standard math programming problem: optimal product mix, scheduling, distribution, capital budgeting, financial planning, facility location, and rating sports teams (ok, this last one is not a standard math programming problem). This focus on providing non-trivial examples for substantial tools is the strength of the book.

Are there things that are missing? Two areas that I find myself teaching students in using spreadsheets is using the tabs to organize data and general spreadsheet management. While there are chapters on pulling in data from text files and the internet as well as a chapter on using auditing tools, there is no chapter on the art of using spreadsheets. The book falls into a common software focused book on working through the functions, even if it does this very well. Therefore, it would not serve as as a book to be used by specialists in data analysis, who need to know the art and purpose of data analysis in addition to applying techniques. But it is good for generalists who have to do analysis as part of their jobs as well as those who need to package analysis to a business audience. And for this purpose, Winston’s book does a good job.

Disclaimer: I received a free electronic copy of Microsoft Excel 2010 Data Analysis and Business Modeling 3rd ed. by Wayne L. Winston through the O’Reilly Press Blogger review program.

I review for the O'Reilly Blogger Review Program

(2) McCoullough and Heiser, 2008, On the accuracy of statistical procedures in Microsoft Excel 2007, Computational Statistics & Data Analysis Vol 52 (10), pp 4570-4578.
Post a Comment