Poisson simultaneous autoregressive analysis of poverty in the Philippines using national household targeting system for poverty reduction (NHTS-PR) data

Date of Publication

2018

Document Type

Bachelor's Thesis

Degree Name

Bachelor of Science in Statistics Major in Actuarial Science

Subject Categories

Physical Sciences and Mathematics

College

College of Science

Department/Unit

Mathematics and Statistics

Abstract/Summary

Creating or finding the most efficient solution to diminish the prevalence of poverty in the Philippines remains as one of the country's major struggles. This paper formulates a spatial model that could aid in poverty reduction using the NHTS-PR 2015 data to identify which indicator variables have significant relationships with poverty count. Given the use of count data for modeling, the Simultaneous Autoregressive (SAR) models were modified to include the Poisson regression approach in the estimation of parameters. The Poisson-SAR models were generated using the backfitting algorithm and were compared with the Ordinary Least Squares (OLS) model for model accuracy. It was found that the Poisson-SARerr model with regional and class dummy variables has the lowest Mean Absolute Percentage Error (MAPE) and provides the most accurate poverty map.

Abstract Format

html

Language

English

Format

Electronic

Accession Number

CDTU017670

Shelf Location

Archives, The Learning Commons, 12F, Henry Sy Sr. Hall

This document is currently not available here.

Share

COinS