Bi-response Spline Smoothing Estimator for Modelling the Percentage of Poor Population and Human Development Index in Papua Province

Abstract

The percentage of the poor population (PPP) and the Human Development Index (HDI) are important indicators to measure the success of a country's development. Papua is the province with the highest PPP and the lowest HDI in Indonesia. Therefore, these two indicators in Papua require special attention and appropriate methods, especially when analyzed with other variables as predictors. One of the predictor variables that affect PPP and HDI is the School Participation Rate (SPR). The bi-response Spline Smoothing estimator is applied to this data because there is a correlation between PPP and HDI and the relationship patterns between the two responses with SPR are unknown. The Model estimation is done by applying the Penalized Weighted Least Square method. The best model for PPP and HDI in Papua is a model with a minimum GCV value 8.4554, this model gives R2=97.63% and MSE= 7.0153. The validation of this model produces MAPE value of 6.57%. These empirical results indicate that the bi-response Spline Smoothing is suitable for modelling the PPP and HDI simultaneously in Papua Province.