Forecasting in international relations: Problems and prospects
Name
[Choucri] 1974 Forecasting in international relations.pdf
Size
1.19 MB
Format
Adobe PDF
Checksum (MD5)
3872ae77158539290eff9d136e016a5f
Author(s)
Choucri, Nazli
Date Issued
1974
Publisher
© Gorden and Breach Science Publishers Ltd.
Citation
Choucri, N. (1974). Forecasting in international relations: Problems and prospects. International Interactions, 1(2), 63–86.
Version
Final published version.
Final published version.
Abstract
The author argues that forecasting is a problem of reasoning, of reducing uncertainty, and of bounded speculation, and she identifies four types of forecasting goals: (1) understanding the unknown (i.e., prophecy), (2) controlling future outcomes, (3) understanding the overall dynamics of a system to appreciate present conditions, and (4) planning for the immediate future. The author then specifies alternative forecasting methodologies, from least to most systematic: normative, exploratory‐projective, model‐based (both statistical and functional), simulation‐based, and artificial intelligence. A forcast may further be anchored in four types of initial conditions: structure, probability, preference, and trends and projections (the most prevalent type today). A forecast may also have various purposes, each with an attendant time frame: retrospective, long range, or short range. Having made a forecast, it may be validated in many ways, including interrogation processes, statistical methods, and comparisons with data. In considering the policy implications of forecasts, the researcher must identify a system's manipulables, the costs of manipulations, and the sensitive points. In conclusion, the author notes some critical imperatives for further developments in international relations forecasting.
Terms of Use
Attribution-NonCommercial-NoDerivs 3.0 United States
Persistent DSpace Link
