The modeling of human development index (HDI) in Papua—Indonesia using geographically weighted ridge regression (GWRR)

In regression model, there are several assumptions which have to be fulfilled, one of is the absence multicollinearity between its independent variables. If occurs, parameter estimation model using method least squares results in unbiased estimators, and even estimators likely large variance. Such variance causes hypothesis testing towards tend accept H0, meaning that coefficient insignificant. One approach deal with multicolinearity biased estimator, ridge regression. Ridge defined as modification by adding a small constant c diagonal elements X’X. The value reflects degree bias estimators. application, fit for discussed article, Human Development Index (HDI) influencing factors. HDI standard based on components human quality life composed three aspects including health, education, living standards. Indonesia has shown an increase last few years, except Papua. Papua was province lowest such factors health care quality, education employment sector, population condition. low 29 regencies/ cities caused difference characteristics geographical conditions spatial units known effect heterogeneity. To heterogeneity, geographically weighted (GWR) applied encounter occurrence multicollinearity, GWRR used. research, data province. predicted (c) appropriate 0.149. From distribution variables significant can constructed.