Estimating household income of the 1998 annual poverty indicator survey on the province of Cavite

• Main Authors: Nemis, Christine Jhoanna S., Salcedo, Madelaine M.
• Format: text
• Language: English
• Published: Animo Repository 2000
• Online Access: https://animorepository.dlsu.edu.ph/etd_bachelors/17000
• Institution: De La Salle University

This paper is about finding the best model for the household income for the province of Cavite, using the 1998 Annual Poverty Indicator Survey. It has a hierarchical structure, wherein the researchers only used two levels: household and domain. Four methods are used to be able to obtain the objective of this paper. It used total or pooled regression and aggregate regression. These two procedures are used for comparison with the two remaining methods to arrive at the best model for household income. Multilevel Analysis is then applied, yet wasn't able to arrive with results since this analysis may not be appropriate for the structure of the data at hand. An alternative procedure was then used, the three-stage regression. In this procedure, it arrived with the best model using Least Squares Estimation procedure for its first level and Binary Logit for the second level and Least Squares again. The derived model showed the factors that contribute significantly with the household income, and these are entrepreneurial activities income, salary or wage, total expenditure and the urbanity of the domain.