Full Citation

Title: Modeling Local Spatial Dependence in Shaping Population Distribution

Citation Type: Working Paper

Publication Year: 2018

ISBN:

ISSN: 1556-5068

DOI: 10.2139/ssrn.3136302

NSFID:

PMCID:

PMID:

Abstract: This study proposes to investigate the effectiveness of modeling local spatial dependence through a conditionally autoregressive process (CAR) to picture the population distribution across space. Following the current literature, the idea is to model individual location preferences by focusing on a selected sample of location determinants but also taking into account spatial dependence in location choices. By exploiting a Bayesian setting, our novelty is to include spatial CAR census random effects that have been found to be effective in enhancing the role of spatial proximity. Our results indicate that to effectively control for the spatial dimension in location choices, one needs to define both a global indicator for space structure and local spatial dependence which has become increasingly important in recent decades.

Url: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3136302

Url: https://www.ssrn.com/abstract=3136302

User Submitted?: No

Authors: Epifani, Ilenia; Ghiringhelli, Chiara; Nicolini, Rosella

Series Title:

Publication Number:

Institution:

Pages:

Publisher Location:

Data Collections: IPUMS NHGIS

Topics: Other, Population Data Science

Countries: