A MULTIPLE REGRESSION TECHNIQUE FOR POPULATION ESTIMATION.

Abstract

The multiple regression population estimation model developed by this paper was designed to meet the criteria of simplicity, accuracy, sensitivity and data availability. The model, as a result of insufficient time and funds for extensive data collections, has not been rigorously tested. Throughout the testing and verification it has received, it has successfully met the above criteria. However, the validity of the model must yet be verified over time and in different cities before any true measure of its worth as a general population estimation model for urban areas will be known. Based upon the testing of the model conducted by the author, the prognoses of its value as an urban area population estimation tool is very good. The model estimates the population of an area at any point in time for which the required input data is available. This is accomplished with a function derived by use of multiple regression. The independent variables used in the function are: elementary school enrollment, number of available housing units and the number of utility meters in service. The coefficients of the model, derived with data from Pittsburgh, Pennsylvania, as of April 1, 1960 are presented. The model, using these coefficients, is tested using data from Pittsburgh, Pennsylvania; Seattle, Washington and Los Angeles, California.

Document Details

Document Type
Technical Report
Publication Date
Jan 01, 1944
Accession Number
AD0605439

Entities

People

  • Edward F. Williams

Organizations

  • University of Pittsburgh

Tags

DTIC Thesaurus Topics

  • Accuracy
  • Availability
  • California
  • Coefficients
  • Continents
  • Geographic Regions
  • North America
  • Pennsylvania
  • Sensitivity
  • United States
  • Urban Areas

Readers

  • Computational Modeling and Simulation
  • Environmental Remediation and Restoration.
  • Women's Health and Cancer Risk Research: African American Women and Pregnancy Outcomes.