Full Citation

Title: Comparing Artificial Neural Network and Cohort-Component Models for Population Forecasts

Citation Type: Conference Paper

Publication Year: 2017

ISBN:

ISSN:

DOI:

NSFID:

PMCID:

PMID:

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 with population projections from traditional cohort-component method (CCM) for counties in Alabama. The evaluation includes forecasts for all 67 counties that offer diversity in terms of population and socioeconomic characteristics. When comparing predicted 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 more accurate results than a basic multi-county ANN model. Only when we use single-county models or proxy for a forecaster's experience and personal judgment with potential economic forecasts, results from ANN models improve. The results indicate the significance of forecaster's experience and judgment for CCM and difficulty, but not impossibility of substituting these insights with available data.

Url: https://paa.confex.com/paa/2017/meetingapp.cgi/Paper/16403

User Submitted?: No

Authors: Riiman, Viktoria; Wilson, Amalee; Milewicz, Reed; Pirkelbauer, Peter

Conference Name: PAA 2017

Publisher Location: Chicago, Illinois

Data Collections: IPUMS NHGIS

Topics: Methodology and Data Collection, Other

Countries: