Ridge regression modeling in overcoming multicollinearity problems in multiple linear regression models (case study: Life expectancy in Maluku Province)

Abstract

Life expectancy rate is one of the indicators used to measure the health status of the population which describes the quality of life. The Life Expectancy Rate of Maluku Province from year to year has always increased. However, there are still disparities or gaps between regions and there are still 7 districts that have a life expectancy below the provincial and national life expectancy rates. This cannot be separated from various influencing factors. So it is necessary to identify the factors that rationally affect the Life Expectancy Rate, these factors tend to have high collinearity, causing multicollinearity cases. If the multicollinearity case is not resolved, it can cause the variance of the parameter estimation results to be large so that it can result in the number of predictor variables that are not significant. To overcome this problem, one of them is by modeling the Life Expectancy Rate by using ridge regression. The results of this study show that the factors that influence the Life Expectancy Rate in Maluku Province are the factor of infant mortality, the factor of toddlers aged 1-4 years who get complete immunization, factors for infants aged 0-11 months who are breastfed for 4-6 months, Education Level factors, Population Literacy Rate factors aged 10 years and over, Labor Force Participation Level factors, Population Growth Rate factors, GRDP factors, and Economic Growth Rate factors. The magnitude of the influence of the independent variables on the Life Expectancy Rate in Maluku Province is 100%.