Viktoria Riiman, Amalee Wilson, Reed Milewicz, P. Pirkelbauer
{"title":"Comparing Artificial Neural Network and Cohort-Component Models for Population Forecasts","authors":"Viktoria Riiman, Amalee Wilson, Reed Milewicz, P. Pirkelbauer","doi":"10.1353/prv.2019.0008","DOIUrl":null,"url":null,"abstract":"Abstract:Artificial neural network (ANN) models are rarely used to forecast population in spite of their growing prominence in other fields. We compare the forecasts generated by ANN long short-term memory models (LSTM) with population projections from the traditional cohort-component method (CCM) for counties in Alabama, USA. The evaluation includes projections for all 67 counties, which are diverse in population and socioeconomic characteristics. When comparing projected values with total population counts from the 2010 decennial census, the CCM used by the Center for Business and Economic Research at the University of Alabama in 2001 produced comparable or better results than a basic multi-county ANN LSTM model. Results from ANN models improve when we use single-county models or proxy for a forecaster’s experience and personal judgment with potential economic forecasts. The results indicate the significance of forecaster’s experience/judgment for CCM and the difficulty, but not impossibility, of substituting these insights with available data.","PeriodicalId":43131,"journal":{"name":"Population Review","volume":null,"pages":null},"PeriodicalIF":0.4000,"publicationDate":"2019-10-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"9","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Population Review","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1353/prv.2019.0008","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q4","JCRName":"DEMOGRAPHY","Score":null,"Total":0}
引用次数: 9
Abstract
Abstract:Artificial neural network (ANN) models are rarely used to forecast population in spite of their growing prominence in other fields. We compare the forecasts generated by ANN long short-term memory models (LSTM) with population projections from the traditional cohort-component method (CCM) for counties in Alabama, USA. The evaluation includes projections for all 67 counties, which are diverse in population and socioeconomic characteristics. When comparing projected values with total population counts from the 2010 decennial census, the CCM used by the Center for Business and Economic Research at the University of Alabama in 2001 produced comparable or better results than a basic multi-county ANN LSTM model. Results from ANN models improve when we use single-county models or proxy for a forecaster’s experience and personal judgment with potential economic forecasts. The results indicate the significance of forecaster’s experience/judgment for CCM and the difficulty, but not impossibility, of substituting these insights with available data.
期刊介绍:
Population Review publishes scholarly research that covers a broad range of social science disciplines, including demography, sociology, social anthropology, socioenvironmental science, communication, and political science. The journal emphasizes empirical research and strives to advance knowledge on the interrelationships between demography and sociology. The editor welcomes submissions that combine theory with solid empirical research. Articles that are of general interest to population specialists are also desired. International in scope, the journal’s focus is not limited by geography. Submissions are encouraged from scholars in both the developing and developed world. Population Review publishes original articles and book reviews. Content is published online immediately after acceptance.