ESTIMATOR KERNEL PADA REGRESI NONPARAMETRIK MENGGUNAKAN PENDEKATAN FUNGSI KERNEL GAUSSIAN
Abstract
The study by using the data to model a state (variable) in the statistical analysis usually requires certain assumptions in order to use the analysis results in accordance with the actual situation. This study uses a nonparametric procedure to estimate a function in which the function does not lead to a certain model of a particular function. The main problem of regression analysis is to determine the shape estimation. One approach that can be used to determine is a kernel estimator with a Gaussian kernel approach. The data used is data that the percentage of women aged 15-49 who have been married according to the last birth attendants in South Sulawesi with the gynecologist predictor variables (), general practitioners (), midwives (), and the response variable () the number of women who have been married according to the last birth attendants. Methods GCV (Generalized Cross Validation) is used to obtain optimal bandwidth that is at , and = 25 with value GCV is . The optimum value is the maximum value of the percentage of women aged who have been married according to the last birth attendants in South Sulawesi.
Moraref