Fourier series semiparametric regression models (Case study: The production of lowland rice irrigation in central Java)

Semiparametric regression model is a regression model where the shape of regression curve consists of a known pattern of parametric components and a smooth (smooth, flawless, slippery) nonparametric component which the pattern is unknown. The approach that used in estimating the nonparametric regression curves, one of which is, the Fourier series estimator. Fourier series estimator is commonly used when a data investigated patterns are not known and there is a tendency of repeating patterns. In the Fourier series estimator, the shape of nonparametric regression curve is assumed unknown and is contained in the space of continuous functions C (0, π). This study aimed to analyze the shape of the estimator of the Fourier series semiparametric regression curve and applying it's to the data production of lowland rice irrigation in Central Java. Case studies are used to model the production of lowland rice irrigation in Central Java with predictor variables harvest area, the use of fertilizers, pesticides, seed, and the use of labor. Modeling aimed to determine the magnitude influence of the predictor variables on the response variable that is the number of production of lowland rice irrigation in Central Java. Modeling the production of lowland rice irrigation in Central Java with Fourier series semiparametric regression produced the coefficient value of determination R² = 0.92. It means that the magnitude influence of the predictor variables on the response variable is 92%. The performance of Fourier series semiparametric regression model was quite good in modeling the production of lowland rice irrigation in Central Java.