No Detail Found!

Estimation of clearness index model via crs, tprs and mars

Author(s)
Özlem Alpu Department of Statistics, Eskisehir Osmangazi University, Eskisehir, TURKEY
Betül Kan Department of Statistics, Anadolu University, Eskisehir, TURKEY
Berna Yazici Department of Statistics, Anadolu University, Eskisehir, TURKEY
Abstract
Nonparametric approach is more flexible than parametric approach in assuming  that  f belongs to a smooth family of functions. Hence, a nonparametric approach does  not require an assumption of linearity. Based on our motivating applications, mainly the  approach to nonparametric regression is used on clearness index of Eskisehir, Turkey.  In this study, cubic regression splines (CRS), thin plate regression splines (TPRS),  multivariate adaptive regression splines (MARS) are defined to explore the shape of the  functional relationship of the data by constructing numerous models for each. Thin plate  regression spline gives the best results for the model in which monthly average daily  extraterrestrial radiation on horizontal surface taken as a parametric component and  monthly average soil temperature, monthly average sunshine hours are taken as  nonparametric components for this data set. 
Publication Details
Page(s) 95-109
DOI DOI not available
Published Journal: Pakistan Journal of Statistics, Volume: 27, Issue: 1, Year: 2011

Request the full text from PASTIC

0

Citations

0

Downloads

93

Views