Pemodelan Fluktuasi Harga Beras di Provinsi Jawa Timur dengan Regresi Logistik Biner dan Regresi Probit Biner

Ramadani, Izza Merry Vitah (2024) Pemodelan Fluktuasi Harga Beras di Provinsi Jawa Timur dengan Regresi Logistik Biner dan Regresi Probit Biner. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5003201010-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 October 2026. Download (2MB) | Request a copy

Abstract

Beras memegang peran sentral dalam kehidupan masyarakat Indonesia, tidak hanya sebagai sumber energi utama tetapi juga dalam menjaga stabilitas ekonomi dan pangan negara. Penelitian ini menganalisis fluktuasi harga beras di Jawa Timur menggunakan regresi logistik biner dan regresi probit biner. Model regresi probit biner dengan backward elimination lebih baik daripada regresi logistik biner berdasarkan AIC dan standar error. Hasil menunjukkan bahwa Luas Panen Padi berpengaruh negatif dan Temperatur Maksimum berpengaruh positif terhadap fluktuasi harga beras. Dalam model dengan prediktor threshold rata-rata harga beras, Inflasi, Luas Panen Padi, dan Curah Hujan signifikan, dengan Inflasi dan Luas Panen Padi berpengaruh negatif, sementara Curah Hujan berpengaruh positif. Dalam model regresi logistik biner, setiap tambahan 1 hektar Luas Panen Padi meningkatkan peluang kenaikan harga beras sebesar 1,000015 kali, dan setiap peningkatan suhu maksimum 1°C meningkatkan peluang kenaikan harga beras sebesar 5,248 kali. Untuk prediktor threshold rata-rata harga beras, Inflasi meningkatkan peluang harga di atas rata-rata sebesar 103,648 kali, Luas Panen Padi sebesar 1,0000076 kali, dan Curah Hujan setiap 1 mm meningkatkan peluang sebesar 1,058 kali. Dalam model regresi probit biner, peningkatan Luas Panen Padi mengurangi probabilitas harga beras naik sebesar 2,104, sementara peningkatan Temperatur Maksimum menaikkan probabilitas sebesar 0,219. Untuk model threshold rata-rata harga beras, Inflasi, Luas Panen Padi, dan Curah Hujan meningkatkan probabilitas harga di atas rata-rata masing-masing sebesar 0,0161; 0,0263; dan 0,0002.
==============================================================================================================================
Rice plays a central role in the lives of Indonesian society, not only as a primary source of energy but also in maintaining the country’s economic and food stability. This study analyzes the fluctuations in rice prices in East Java using binary logistic regression and binary probit regression. The probit regression model with backward elimination is found to be better than the binary logistic regression model based on AIC and standard errors. The results indicate that Paddy Harvest Area has a negative impact, while Maximum Temperature has a positive impact on rice price fluctuations. In the model with the average rice price threshold predictor, Inflation, Paddy Harvest Area, and Rainfall are significant, with Inflation and Paddy Harvest Area having negative effects, and Rainfall having a positive effect. In the binary logistic regression model, each additional hectare of Paddy Harvest Area increases the likelihood of a rise in rice prices by 1,000015 times, and each 1°C increase in Maximum Temperature increases the likelihood of a rise in rice prices by 5.248 times. For the average rice price threshold predictor model, Inflation increases the likelihood of prices being above average by 103,648 times, Paddy Harvest Area by 1,0000076 times, and each 1 mm increase in Rainfall raises the likelihood by 1,058 times. In the binary probit regression model, an increase in Paddy Harvest Area reduces the probability of a rise in rice prices by 2,104, while an increase in Maximum Temperature raises the probability by 0,219. For the average rice price threshold predictor model, Inflation, Paddy Harvest Area, and Rainfall increase the probability of prices being above average by 0,0161; 0,0263, and 0,0002, respectively

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Beras, Fluktuasi, Jawa Timur, Regresi Logistik Biner, Regresi Probit Biner, Binary Logistic Regression, Binary Probit Regression, East Java, Fluctuations, Rice
Subjects:	Q Science > QA Mathematics > QA278.2 Regression Analysis. Logistic regression
Divisions:	Faculty of Science and Data Analytics (SCIENTICS) > Statistics > 49201-(S1) Undergraduate Thesis
Depositing User:	Izza Merry Vitah Ramadani
Date Deposited:	08 Aug 2024 07:28
Last Modified:	27 Aug 2024 07:02
URI:	http://repository.its.ac.id/id/eprint/114649

Actions (login required)

View Item