Open AccessPENGUJIAN HIPOTESIS SIMULTAN MODEL REGRESI NONPARAMETRIK SPLINE TRUNCATED DALAM PEMODELAN KASUS EKONOMI
Author(s) -
Andrea Tri Rian Dani,
Narita Yuri Adrianingsih,
Alifta Ainurrochmah
Publication year - 2020
Publication title -
jambura journal of probability and statistics
Language(s) - English
Resource type - Journals
ISSN - 2722-7189
DOI - 10.34312/jjps.v1i2.7755
Subject(s) - mathematics , statistics , nonparametric statistics , estimator , nonparametric regression , spline (mechanical) , goodness of fit , regression analysis , econometrics , engineering , structural engineering
The pattern in a relationship between the response variable and the predictor variable can be known and some cannot be known. In determining the unknown pattern of relationships, nonparametric regression approaches can be used. The nonparametric regression approach is very flexible. One of the most frequently used nonparametric regression approaches is the truncated spline. Truncated splines are polynomial pieces that are segmented and continuous. The purpose of this study is to obtain the best estimator model in the Gini Ratio case against the variables suspected of influencing it, then perform simultaneous hypothesis testing on the nonparametric regression model. The criteria for the goodness of the model use the GCV and R2 values. In the case modeling of the District / City Gini Ratio in East Java Province using a nonparametric regression approach, it was found that the truncated spline estimator with 3 knots points gave quite good results. This is indicated by the coefficient of determination of the truncated spline estimator, which is 84.76%. Based on the results of simultaneous testing, it was found that the open unemployment rate, the percentage of poor people and the rate of economic growth simultaneously had an influence on the Gini Ratio.