PEMODELAN INDEKS PEMBANGUNAN MANUSIA DI JAWA TENGAH MENGGUNAKAN METODE REGRESI RIDGE DAN REGRESI STEPWISE

The Human Development Index (HDI) is an important indicator in measuring the success of national development. Central Java with a high population can be considered as obstacle and driver To find out factors that affect HDI, it necessary to make model. One statistical methods used multiple linear regression analysis. However, modeling there are assumptions must met, namely linearity, normality, homoscedasticity, non-autocorrelation, non-multicollinearity. If non-multicollinearity assumption not then another alternative needed estimate parameters. Several ridge stepwise methods. best model selection done by looking at smallest Mean Square Error (MSE) value. In this study, were applied HDI data 2021 influence it, life expectancy birth, expected years schooling, average length per capita expenditure, percentage poor people, unemployment open. Based on Variance Inflation Factor (VIF) value more than 10, concluded multicollinearity violation. Modeling produces model, MSE R square 0,99 indicates included criteria for strong