Estimasi Kurva Regresi Nonparametrik Heteroskedastisitas Spline (Studi Kasus Berat Badan Balita di Kecamatan Kerambitan, Bali)

Hendayanti, Ni Putu Nanik (2015) Estimasi Kurva Regresi Nonparametrik Heteroskedastisitas Spline (Studi Kasus Berat Badan Balita di Kecamatan Kerambitan, Bali). Masters thesis, Institut Technology Sepuluh Nopember.

[img]
Preview
Text
1313201018-Master Theses.pdf - Published Version

Download (2MB) | Preview

Abstract

Pendekatan nonparametrik merupakan metode estimasi yang tidak terikat asumsi bentuk kurva tertentu. Pendekatan regresi nonparametrik yang sering digunakan adalah spline. Spline memiliki kemampuan yang sangat baik untuk menangani data yang perilakunya berubah-ubah pada sub-sub interval tertentu. Pada regresi nonparametrik, estimator spline sangat tergantung pada titik knot optimal, dimana pemilihan titik knot optimal berdasarkan nilai GCV (Generalized Cross Validation) yang minimum. Dalam penelitian ini, penulis mengestimasi kurva gˆ dengan menggunakan optimasi Likelihood dan mengkontruksi selang kepercayaan untuk kurva regresi g dengan pendekatan spline menggunakan Pivotal Quantity. Model regresi yang diteliti adalah model regresi nonparametrik spline heteroskedastisitas. ===================================================================================================== Nonparametric approachs are an estimation methods not tied on particular shape of the curve assumptions. The most frequently used of nonparametric regression approach is spline. Spline has an excellent ability to handle data that behavior change in sub-specified interval. In nonparametric regression, spline estimator depends on the point of optimal knots, which is the selection of the optimal knots based on the value of GCV (Generalized Cross Validation) minimum. In this study, gˆ curve is estimated using Likelihood optimization and confidence intervals for the regression curve by spline approach is constructed using Pivotal Quantity. The regression models of interest is heteroskedasticity spline nonparametric regression models. Therefore, it is necessary to give a weight to overcome the heteroskedasticity

Item Type: Thesis (Masters)
Additional Information: RTSt 519.536 Hen e
Uncontrolled Keywords: Nonparametric Regression, Spline, Heteroskedasticity, Pivotal Quantity
Subjects: Q Science > QA Mathematics > QA278.2 Regression Analysis
Divisions: Faculty of Mathematics and Science > Statistics > 49101-(S2) Master Thesis
Depositing User: Mr. Tondo Indra Nyata
Date Deposited: 04 Jun 2018 02:36
Last Modified: 04 Jun 2018 02:36
URI: https://repository.its.ac.id/id/eprint/51977

Actions (login required)

View Item View Item