School of Business

Comparative Forecasting-Hindsighting the Czechoslovakian Invasion

View Full Record
Description: It is possible to utilize 2 dissimilar information processing techniques to forecast the same phenomenon. The Bayesian method rests heavily on subjective information, while event-pattern recognition is objective in nature, uses predetermined indicators, and is dependent on selected data flows. Both of these methods may be used to forecast a variety of phenonmena, and this research retrospectively tests them for their ability to forecast the August, 1968, Soviet invasion of Czechoslovakia. It is apparent that objective and subjective methods may be used to generate combined, or at least cross-validated forecasts. Comparative forecasting thus may be used productively for international political analysis. It is illustrated how, and at what costs, it may be utilized. Benefits would accrue from quantitative political scientists and Bayesians working together to enhance forecasting capability.
Language: English
Format: Article