Inside the Crystal Ball: How to Make and Use Forecasts

Chapter 2The Art and Science of Making and Using Forecasts

Ultimately there's no alternative to judgment—you can never get the answers out of some model.

—Lawrence Summers, “What in the World Happened to Economics?” Fortune, March 15, 1999, p. 66

What are the overall philosophies, habits, and techniques that have enabled me to build a reasonably successful track record during my almost four-decade career in professional forecasting? What have I learned from both the successes and failures of the many economists and securities analysts with whom I have worked? In addressing these questions, my initial focus is on what it takes to achieve good long-run forecasting performance for a wide range of variables. Specific techniques and issues are addressed in subsequent chapters.

Statistics courses are a standard element of every business school curriculum, so most business and financial forecasters share a similar academic training in the discipline. Performance among similarly educated professionals can vary widely, however. With this in mind, I believe it is important to first address one's overall approach to, or philosophy of, forecasting and work habits before assessing the many methodological challenges encountered in making forecasts. In my experience, having an edge in forecasting can have more to do with these nontechnical issues than with a specific educational background.

Judgment Counts More Than Math

Statistics and economics texts can be intimidating if math is not one's ...