Comparing Artificial Neural Network and Cohort-Component Models for Population Forecasts
- * Population Review
- * Vol. 58, Number 2, 2019
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.
Population Review
Type: Article, pp. 100-116
Comparing Artificial Neural Network and Cohort-Component Models for Population Forecasts
Authors: Viktoria Riiman, Amalee Wilson, Reed Milewicz, Peter Pirkelbauer
Affiliations: Center for Business and Economic Research, University of Alabama (Riiman); Computer Science Department, Stanford University (Wilson); Center for Computing Research, Sandia National Laboratories (Milewicz); Center for Applied Scientific Computing, Lawrence Livermore
National Laboratory (Pirkelbauer)
Corresponding author/address: Viktoria Riiman, Center for Business and Economic Research, University of Alabama; email: [email protected]
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.
Keywords
Population forecast, population projection, artificial neural networks, cohort-component method
Data availability statement: The data that support the findings of this study are openly available in OSF
at https://osf.io/89WFN/, identifier DOI 10.17605/OSF.IO/89WFN.
Acknowledgement: We are grateful for the constructive comments from Ron Prevost. This work was run on computer equipment purchased using NSF grants CNS-0821497 and CNS-1229282.
© 2019 Sociological Demography Press
MLA
Riiman, Viktoria, et al.
Comparing Artificial Neural Network and Cohort-Component Models for Population Forecasts.
Population Review, vol. 58 no. 2, 2019.
Project MUSE, doi:10.1353/prv.2019.0008.
APA
Riiman, V., Wilson, A., Milewicz, R., & Pirkelbauer, P. (2019).
Comparing Artificial Neural Network and Cohort-Component Models for Population Forecasts.
Population Review 58(2), doi:10.1353/prv.2019.0008.
Chicago
Riiman, Viktoria, Amalee Wilson, Reed Milewicz, and Peter Pirkelbauer.
Comparing Artificial Neural Network and Cohort-Component Models for Population Forecasts.
Population Review 58, no. 2 (2019) doi:10.1353/prv.2019.0008.
Endnote
TY – JOUR T1 –
Comparing Artificial Neural Network and Cohort-Component Models for Population Forecasts
A1 – Riiman, Viktoria A1 – Wilson, Amalee A1 – Milewicz, Reed A1 – Pirkelbauer, Peter JF – Population Review VL – 58 IS – 2 PY – 2019 PB – Sociological Demography Press SN – 1549-0955 UR – https://muse.jhu.edu/article/736444 N1
Vol. 58, Number 2, 2019 ER