Suchergebnisse

The nature & use of "complex" models for forecasting & policy simulation are analyzed on theoretical & empirical-performance grounds. While the accuracy of complex models in forecasting trends in such fields as economic & energy is undistinguished, their special virtues -- preserving counterintuitive results & representing subsystem interdependence -- could be used to better advantage than current practice permits. Suggestions for such improvements, through more diversified models, microprocess models in addition to the typical macro models, a mix of mechanically & judgmentally operated models, & the modeling of policy response, are reviewed. 26 References. Modified HA.