APLIKASI SPLINE ESTIMATOR TERBOBOT

Main Author: Budiantara, I Nyoman; Fakultas Matematika dan Ilmu Pengetahuan Alam, Jurusan Statistika, Institut Teknologi 10 November Surabaya
Format: Article info application/pdf eJournal
Bahasa: eng
Terbitan: Institute of Research and Community Outreach - Petra Christian University , 2004
Subjects:
Online Access: http://jurnalindustri.petra.ac.id/index.php/ind/article/view/16001
http://jurnalindustri.petra.ac.id/index.php/ind/article/view/16001/15993
Daftar Isi:
  • We considered the nonparametric regression model : Zj = X(tj) + ej, j = 1,2,…,n, where X(tj) is the regression curve. The random error ej are independently distributed normal with a zero mean and a variance s2/bj, bj > 0. The estimation of X obtained by minimizing a Weighted Least Square. The solution of this optimation is a Weighted Spline Polynomial. Further, we give an application of weigted spline estimator in nonparametric regression. Abstract in Bahasa Indonesia : Diberikan model regresi nonparametrik : Zj = X(tj) + ej, j = 1,2,…,n, dengan X (tj) kurva regresi dan ej sesatan random yang diasumsikan berdistribusi normal dengan mean nol dan variansi s2/bj, bj > 0. Estimasi kurva regresi X yang meminimumkan suatu Penalized Least Square Terbobot, merupakan estimator Polinomial Spline Natural Terbobot. Selanjutnya diberikan suatu aplikasi estimator spline terbobot dalam regresi nonparametrik. Kata kunci: Spline terbobot, Regresi nonparametrik, Penalized Least Square.