{"title":"Forecasting Political Events","authors":"Michael C. Horowitz","doi":"10.1093/oxfordhb/9780190634131.013.16","DOIUrl":null,"url":null,"abstract":"Forecasting political events is a critical activity for social scientists. Forecasting can help test competing theories, let researchers grapple with the true substantive effects of their models, and bridge the gap between academia and the policy world. Forecasting is an academic activity with direct relevance for policymakers. Yet, a variety of cognitive biases can make forecasting challenging, even for experts. Despite these limitations, interest in forecasting is growing. This chapter describes several different approaches to forecasting political events, especially international political events. These methods include game theory, machine learning, statistical analysis, and event data algorithms. Recent research also suggests the way models drawing on the wisdom of the crowds, forecasting teams, and prediction markets can generate large improvements in accuracy when forecasting geopolitical events. All have strengths and weaknesses, given the inherent uncertainty that exists in the political world.","PeriodicalId":106674,"journal":{"name":"The Oxford Handbook of Behavioral Political Science","volume":"74 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2021-01-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"1","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"The Oxford Handbook of Behavioral Political Science","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1093/oxfordhb/9780190634131.013.16","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 1
Abstract
Forecasting political events is a critical activity for social scientists. Forecasting can help test competing theories, let researchers grapple with the true substantive effects of their models, and bridge the gap between academia and the policy world. Forecasting is an academic activity with direct relevance for policymakers. Yet, a variety of cognitive biases can make forecasting challenging, even for experts. Despite these limitations, interest in forecasting is growing. This chapter describes several different approaches to forecasting political events, especially international political events. These methods include game theory, machine learning, statistical analysis, and event data algorithms. Recent research also suggests the way models drawing on the wisdom of the crowds, forecasting teams, and prediction markets can generate large improvements in accuracy when forecasting geopolitical events. All have strengths and weaknesses, given the inherent uncertainty that exists in the political world.