SIGNIFICANT INDICATORS OF LOW COST HOUSING DEMAND: COMPARISON BETWEEN RESULTS OBTAINED FROM PRINCIPAL COMPONENT ANALYSIS, BACK ELIMINATION AND REGRESSION ANALYSIS METHODS
Keywords:principal component analysis, back elimination method, regression method, low cost housing demand, Indicator of low cost housing, Malaysia
AbstractThis paper has reported comparison between Principal Component Analysis (PCA), Back Elimination Method (BEM) and Regression Method. These techniques were applied by using statistical software package SPSS 13.0. For the purpose of comparison, all the methods were tested on nine prime indicators of low cost housing demand which include population growth, birth rate, mortality rate, inflation rate, unemployment rate, GDP (gross domestic product), housing stock, household income and poverty rate. Data for the indicators was obtained from ministry of housing for low cost housing demand in Gombak District. From analysis it was found that PCA method had identified three significant indicators for low cost housing demand that is GDP/Capita in Selangor, housing stock and mortality baby rate. BEM had identified four significant indicators that is inflation rate, GDP/Capita in Selangor, Poverty Rate and Housing Stock. While, regression method identified only one significant indicator that is poverty rate. From these findings it can be concluded that BEM is the best method in determining significant indicators as compared to PCA and regression method. These finding will help the researcher in adopting suitable method for determining significant indicators in any field.
How to Cite
Open access licenses
Open Access is by licensing the content with a Creative Commons (CC) license.
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.