MODELING COMMUTE TIME IN NEW YORK CITY: A SPATIAL ECONOMETRIC APPROACH

Open Access
Author:
Berkman, Benjamin Jacob
Area of Honors:
Economics
Degree:
Bachelor of Science
Document Type:
Thesis
Thesis Supervisors:
  • Amanda Mandzik, Thesis Supervisor
  • Russell Chuderewicz, Honors Advisor
Keywords:
  • income
  • commute time
  • geographically weighted regression
  • urban economics
  • arcgis
Abstract:
This paper seeks to determine whether income as well as other socioeconomic variables are strong determinants of commute time in New York City. Using both linear and spatial regression analysis, we test and visualize United States Census Bureau data of commute time, income, and other select economic indicators. We begin with a review of the wealth of literate on the relationship between income and commute time, then highlight more specific studies conducted in New York City. We next apply ordinary least squares regression analysis, and Geographically Weighted Regression tools via geographic information system software to understand whether the projected results based on previous research match the observed outcomes in the parameters of this study. We find that while income doesn’t explain commute time linearly, it does so in many cases spatially – localized regressions of neighboring census tracts do effectively model commute time. We further conclude that education rate, the non-minority rate, and population density are relatively strong determinants of commute time in New York City at the census tract level.