Perbandingan FPCA dan FPCR dalam Pemodelan Data Curah Hujan Kabupaten Indramayu

Achi Rinaldi(1),
(1) , 


  This research was aimed to modelling rainfall in Indramayu based on the information from Sukadana rainfall station. Modelling rainfall data were very usefull for some prediction in the future and for government policy which can prevent some disaster like flood. Statistics methods in this paper were Principal Component Analysis (PCA), Principal Component Regression (PCR), Functional Principal Component Analysis (FPCA), and Functional Principal Component Regression (FPCR).

Principal Component Regression (PCR) perform a good result compare with Functional Principal Component Regression (FPCR), but without basis function the both method are equal. R2 value for PCR was 50.13 the same with  FPCR without basis function.

RMSEP value and correlation were 77.55 and 0.91 and it showed that PCR was better than FPCR with bspline basis function and fourrier which the value were 110.48 and 0.81, also 136,06 and 0.1. For general FPCR model without basis function was showed the best performance especially FPCR central model that give prediction result for rainfall data was better than the other model. 

Full Text:


Article Metrics

 Abstract Views : 31 times
 PDF Downloaded : 50 times


  • There are currently no refbacks.

View MyStat

Creative Commons License

This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.

Gradien Indexed by :

Tidak ada teks alternatif otomatis yang tersedia.Find in a library with WorldCat