Kartika Fitriasari
Statistika ITS

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

Weight Estimation Using Generalized Moving Average Jerry D. T. Purnomo; I.N. Budiantara; Kartika Fitriasari
IPTEK The Journal for Technology and Science Vol 19, No 4 (2008)
Publisher : IPTEK, LPPM, Institut Teknologi Sepuluh Nopember

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.12962/j20882033.v19i4.140

Abstract

Estimation of regression curve usually conducted using three methods; parametric method, non-parametric method, and semi-parametric method. Non-parametric method has several techniques, which are histogram, kernel, and spline. From various types of spline techniques, weighted parsial spline is developed to solve heterokedasticity problem, this is due to the inability of original partial spline model in handling the heterokedasticity problem. Different techniques are used in choosing the weighted criteria, one of the technique is Generalized Moving Average (GMA). Study about the amount of electricity power loss in PT. PLN East Java Province, North Surabaya Region, resulted that there was a tendency of heterogeneous variance.Using weighted partial spline model with GMA method give better result than original partial spline model. This finding indicates the model of weighted partial spline using GMA method is better than original partial spline model in explaining the heterogeneity of variance.