Penerapan Model Geographically Dan Temporally Weighted Regression Pada Kecelakaan Lalu Lintas
Program Studi Statistika, FMIPA Universitas Tanjungpura, Naomi Nessyana Debataraja, Dadan Kusnandar, Program Studi Statistika, FMIPA Universitas Tanjungpura, Riani Mahalalita, Program Studi Statistika, FMIPA Universitas Tanjungpura, Nurfitri Imro’ah, Program Studi Statistika, FMIPA Universitas Tanjungpura · Jurnal Siger Matematika · 2021
Geographically and temporally weighted regression (GTWR) is a model that is used to deal with instability in data both spatially and temporally and to produce local parameters. In this paper, The GTWR model is used to analyze the factors that are thought to significantly influence the number of traffic accidents in Mempawah Regency. The data used in this study came from 8 districts with the variables used were the number of traffic accidents, the number of population (gender ratio, length of damaged road conditions, and percentage of adolescence. The parameter estimation of the GTWR model was obtained using the weighted least square (WLS) method. The optimal bandwidth selection uses the Cross-Validation (CV) method and the weighting used is the Fixed bisquare function. The results of the analysis show that using the GTWR model, it was found that only the population size variable significantly affected the number of traffic accidents in all locations in Mempawah Regency from 2015 to 2018. The GTWR model was known to be better than the multiple regression model because it produced smaller AIC and RSS values and a larger R-square value.