ABSTRACT
INTRODUCTION Forecasting is a very serious activity in economics, involving a great deal of effort and money to produce them, and some of these forecasts attract a lot of attention with the press. It is probably fair to say that virtually every forecasting method that has ever been proposed has been applied, or at least carefully considered for use with economic data. I believe that this is true just for stock market data alone, for obvious reasons. It is worth remembering that economics is a decision science, with an objective of trying to explain the decisions of the agents and institutions that make up the economy. Forecasts, or the close cousins called “expectations,” are important inputs into these decisions. For example, when a company is considering a large investment in a new factory or piece of machinery, its decision is partly based on a sequence of forecasts of the cash flow that will arise from the investment.
