Simulations Study Combined Estimator Fourier Series and Spline Truncated in Multivariable Nonparametric Regression

9th Annual Basic Science International Conference 2019

自然科学（总论）

Wayan Sudiarsa, I.^1

Department of Mathematics, Faculty of Mathematics and Natural Sciences, Institut Keguruan Dan Ilmu Pendidikan (IKIP) PGRI, Bali, Indonesia^1

关键词: Combined estimators;    Multi variables;    Non-parametric regression;    Penalized least-squares;    Regression function;    Sample sizes;    Simulation data;    Theoretical research;   

Others  :  https://iopscience.iop.org/article/10.1088/1757-899X/546/5/052074/pdf
DOI  :  10.1088/1757-899X/546/5/052074

学科分类：自然科学（综合）

来源: IOP

【 摘 要 】

The problems one of that appeared in everyday life is how to explain the shape of the relation pattern between a response with some predictor. Regression analysis is a statistical method used to estimate the relation pattern between response (y) with predictor (x). Development of nonparametric regression of Fourier series involving multiple predictor, has been more developed for predictors of the same pattern. We need to develop different estimator for different predictors of multivariable nonparametric regression. Theoretical research will be focused on estimator form, and applied to simulation data. The estimation of the combined regression function of the Fourier and spline truncated sequence is obtained through the optimization of penalized least square (PLS). Based on the simulation result obtained that the larger the sample size and the smaller the size of the variance, it will result in better estimation value of parameter and knot.