Smoothing Spline Estimator in Nonparametric Regression (Application: Poverty in Papua Province)

Abstract

Three estimates were obtained in estimating the regression curve, namely estimation of parametric regression, nonparametric regression and semiparametric regression. The most popular nonparametric regression option is smoothing spline. The advantage of smoothing spline is that it can use variable data at certain sub intervals, so this model needs to find its own data estimation. Smoothing Spline allows characters to function smoothly. In everyday life, data patterns are often found to change at certain sub-intervals, one of which is poverty data in Papua Province. Papua Province is ranked first in the percentage of poor people in Indonesia. The best nonparametric Smoothing Spline regression model for the poverty model in Papua Province with a generalized cross validation (GCV) value of 92.77 and R 2 =99.99%.

Comparison of Local Polynomial Regression and Arima in Predicting the Number of Foreign Tourist Visits to Indonesia

“…According to [15], there are several types of kernel functions, namely Kernel Uniform, Kernel Triangular, Kernel Epanechnikov, and Kernel Gaussian.…”

Development of Semiparametric Path Analysis Modeling Truncated Spline: Determinants of Increased Regional Economic Growth

“…In the nonparametric regression approach, the form of the estimated relationship pattern model is determined based on the existing data pattern. With the Spline function technique, the pattern of associations between responses and unknown predictors can be inferred [6], [7], [8] The spline method in nonparametric regression can be found in various forms, namely smoothing spline and truncated spline, where both spline methods use different coefficients in function estimation. The truncated spline method uses knot points, while the smoothing spline method uses smoothing coefficients, where this difference causes the optimization to obtain an estimator that is also different.…”