Open AccessModelling rice production in Central Java using semiparametric regression of local polynomial kernel approach
Author(s) -
Tiani Wahyu Utami,
Alan Prahutama,
Abdul Karim,
A. R. F. Achmad
Publication year - 2019
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1217/1/012108
Subject(s) - polynomial regression , semiparametric regression , nonparametric regression , regression analysis , mathematics , java , parametric statistics , kernel regression , statistics , nonparametric statistics , semiparametric model , kernel (algebra) , regression , kernel smoother , local regression , econometrics , computer science , kernel method , artificial intelligence , radial basis function kernel , discrete mathematics , support vector machine , programming language
Indonesia is an agricultural country with rice as one of the staple foods. Production of rice in the province of Central Java is the highest in Indonesia. The purpose of this study was to model rice production in 31 districts / cities in Central Java Province using semiparametric regression. Semiparametric regression is a combination of parametric and nonparametric regression. Parametric regression curves have a patterned, for example linear, quadratic, and cubic. Nonparametric regression has a smooth curve of the unknown pattern, so in this case required smoothing technique used to smooth curves that one of them is the local polynomial kernel approach and the election of bandwidth the optimal using method Generalized Cross Validation (GCV). Variables used in the study of the production of rice as the response variable, while the predictor variables that harvested area and rainfall. The data used are secondary data from the official website of Central Bureau of Statistics (BPS) of Central Java. Based on the results obtained by applying the model the optimal bandwidth values is 0.43 and polynomial order p = 2 when the minimum GCV so the results of the estimation model R 2 is 0.968