USE OF GCV METHOD IN FORMING ESTIMATORS FOURIER SERIES AND ITS APPLICATIONS

Abstract

Linear regression model with function curve regression and error are normally distributed with a mean of zero and a deviation of standard sigma square. The problems that arise is how form estimate from function curve regression. For estimate function curve regression, there are two approaches that can be taken used that is approach parametric and approximation nonparametric. Approach nonparametric done if there is no assumption form curve regression. Function curve regression only assumed loaded in a room dimensionless function​ not up to. Research this aiming for to study series estimator form fourier in regression model nonparametric. Next under review selection of optimal lambda smoothing parameters with CV and GCV methods. Form of series estimator fourier in regression model nonparametric bulk data rain Cilacap Central Java in the month January 2010 – Dec 2022 is . Next under review selection of optimal lambda smoothing parameters if k = 5 then lambda value = 0.001584 and CV value = 0.0052237 and GCV = 0.0005400.