Regresi Ridge Robust untuk Pemodelan Luas Panen Padi di Kabupaten Gresik dengan Indikator Iklim

Handoko, Anggun Nurfitriani (2020) Regresi Ridge Robust untuk Pemodelan Luas Panen Padi di Kabupaten Gresik dengan Indikator Iklim. Other thesis, Institut Teknologi Sepuluh Nopember.

[thumbnail of 06211640000106-Undergraduate_Thesis.pdf]
Preview
Text
06211640000106-Undergraduate_Thesis.pdf

Download (2MB) | Preview

Abstract

Kabupaten Gresik memiliki produktivitas tanaman padi tertinggi kedua setelah Kota Malang. Pada sektor pertanian, khususnya padi, iklim berpengaruh pada luas area panen, produktivitas, dan kualitas hasil panen. Identifikasi hubungan antara luas panen dan indikator iklim dibutuhkan untuk memodelkan hubungan dari kedua variabel ini. Adanya keragaman curah hujan di wilayah Kabupaten Gresik dan luas panen padi serta dependensi antar data curah hujan menyebabkan amatan outlier dan multikolinearitas. Penelitian ini digunakan metode robust untuk mengatasi amatan outlier dan metode ridge untuk mengatasi masalah multikolinearitas. Hasil penelitian menunjukkan bahwa luas panen padi pada periode pertama terdapat pengamatan outlier, sehingga model optimum adalah regresi robust S-Estimation. Sedangkan, untuk periode kedua dan ketiga yang terdapat amatan outlier dan multikolinearitas, model terbaik diperoleh melalui penggabungan antara metode ridge dan robust, yaitu Regresi Ridge Robust S-Estimation untuk periode kedua dan Regresi Ridge Robust M-Estimation untuk periode ketiga. Dari model yang optimum diperoleh hasil estimasi untuk luas panen padi tahun 2018 dengan akurasi 99,17 persen.
====================================================================================================================
Gresik has the second highest productivity in paddy after Malang. In the agricultural sector, especially paddy, the climate affects the harvested area, productivity, and the quality of the crop. Identification the relation between harvests area and climate indicator is required for modeling the relationship of both variables. However, the rainfall diversity in Gresik and harvested area data as well as the correlations between rainfall data caused outlier cases and multicollinearity cases. In this study, robust regression applied to resolve the outliers and ridge regression to overcome the multicollinearity. The result of study indicates that the harvested area in the first period that have outliers in it, was better to estimated by the robust S-Estimation regression method. Whereas, for the second and third periods that contain outliers and multicollinearity case, the best found from combining the method between ridge and robust regression, these are Ridge Robust S-Estimation Regression for the second period and Ridge Robust M-Estimation Regression for the third period. From the best model the estimation results obtained for the harvested area in 2018 with an accuracy of 99.17 percent.

Item Type: Thesis (Other)
Additional Information: RSSt 519.536 Han r-1 • Handoko, Anggun Nurfitriani
Uncontrolled Keywords: Harvested Area of Paddy, Multicolinierity, Ridge, Robust, Ridge Robust, Luas Panen Padi, Multikolinieritas, Outlier, Ridge, Robust, Ridge Robust
Subjects: S Agriculture > S Agriculture (General)
S Agriculture > S Agriculture (General) > S600.7.R35 Rain and rainfall
Divisions: Faculty of Mathematics and Science > Statistics > 49201-(S1) Undergraduate Thesis
Depositing User: Anggun Nurfitriani Handoko
Date Deposited: 27 Aug 2020 08:00
Last Modified: 15 Dec 2023 11:09
URI: http://repository.its.ac.id/id/eprint/80975

Actions (login required)

View Item View Item