Economic forecasts by industry type: comparisons and statistical tests

Abstract

In this article we group together and compare economic forecasts by industry type. That is, corporate forecasts of GDP and CPI are compiled by industry for the following: financial institutions, economic consultants, auto manufacturers, business associations, universities, and insurance companies. Error statistics, such as the MAE and RMSE, are calculated over the 1992 to 1995 forecasting range. Observing the rankings of industry forecasts based on error statistics might lead one to conclude that some industries are more proficient at forecasting the economy than others. Indeed, we might infer that intentional bias may be a component of the forecasts provided by some industries. This work, however, provides a number of statistical tests to determine whether there are real differences in forecasts by industry type, or whether the differences are merely matters of chance.