PERAMALAN HARGA GABAH KERING PANEN DI INDONESIA MENGGUNAKAN METODE HYBRID AUTOREGRESSIVE INTEGRATED MOVING AVERAGE EXOGENOUS (ARIMAX) – FEED FORWARD NEURAL NETWORK (FFNN)

TRI ZAHROTUN, WAHYUNINGSIH (2022) PERAMALAN HARGA GABAH KERING PANEN DI INDONESIA MENGGUNAKAN METODE HYBRID AUTOREGRESSIVE INTEGRATED MOVING AVERAGE EXOGENOUS (ARIMAX) – FEED FORWARD NEURAL NETWORK (FFNN). Sarjana / Sarjana Terapan (S1/D4) thesis, Universitas Muhammadiyah Semarang.

Preview	Text 0. COVER.pdf Download (530kB) | Preview
Preview	Text 5. ABSTRAK.pdf Download (537kB) | Preview
Preview	Text BAB 1 PENDAHULUAN.pdf Download (1MB) | Preview
Preview	Text BAB 2 TINJAUAN PUSTAKA.pdf Download (1MB) | Preview
	Text BAB 3 METODOLOGI PENELITIAN.pdf Restricted to Repository staff only Download (847kB) | Request a copy
	Text BAB 4 HASIL DAN PEMBAHASAN.pdf Restricted to Repository staff only Download (1MB) | Request a copy
Preview	Text BAB 5 KESIMPULAN DAN SARAN.pdf Download (683kB) | Preview
Preview	Text DAFTAR PUSTAKA.pdf Download (722kB) | Preview
	Text LAMPIRAN.pdf Restricted to Repository staff only Download (871kB) | Request a copy
	Text FULL TEXT.pdf Restricted to Repository staff only Download (2MB) | Request a copy

Abstract

ABSTRAK Wahyuningsih, Tri Zahrotun, 2022, Peramalan Harga Gabah Kering Panen di Indonesia Menggunakan Metode Hybrid Autoregressive Integrated Moving Average Exogenous (ARIMAX)-Feed Forward Neural Network (FFNN). Skripsi, Program Studi Statistika, Universitas Muhammadiyah Semarang. Pembimbing: I. M.Al Haris, S.Si, M.Si., II. Indah Manfaati Nur, S.Si, M.Si. Indonesia merupakan salah satu negara agraris yang mengandalkan sektor pertanian terutama tanaman padi. Permintaan tanaman padi atau gabah mengalami fluktuasi selama masa pandemi Covid-19 sehingga menyebabkan harga gabah mengalami ketidakstabilan. Perubahan harga gabah ini memiliki sifat data linier, selain terdapat pola linier juga terdapat pola non linier sehingga dilakukan peramalan menggunakan metode Hybrid ARIMAX-FFNN dengan variabel dummy Covid-19 yang mempunyai pengaruh signifikan terhadap data harga gabah panen (GKP). Model Hybrid ARIMAX-FFNN terbaik pada data Harga GKP di Indonesia yaitu gabungan ARIMAX (0,1,0) (1,1,0)^12 dan residualnya dimodelkan kembali dengan FFNN menggunakan algoritma resilient Backpropagation. Pemodelan dengan FFNN diperoleh arsitektur terbaik yaitu 12-13-1 (12 neuron input,13 neuron hidden, 1 neuron output). Hasil peramalan harga GKP di Indonesia pada bulan Juni 2022-Mei 2023 mengalami fluktuasi pada setiap bulannya. Harga GKP di Indonesia terendah terjadi pada bulan Juli 2022 sebesar Rp.4310,93 dan tertinggi terjadi pada bulan Januari 2023 sebesar Rp. 5009,67. Hasil prediksi model Hybrid ARIMAX-FFNN diperoleh nilai MAPE sebesar 1.75%. Kata Kunci : Covid-19, Hybrid ARIMAX-FFNN, Harga GKP, Peramalan. ABSTRACT Wahyuningsih, Tri Zahrotun, 2022, Forecasting the Price of Harvested Dry Grain in Indonesia Using the Hybrid Autoregressive Moving Average Exougenous (ARIMAX)-Feed Forward Neural Network (FFNN) Method. Thesis, Statistics Study Program, University of Muhammadiyah Semarang. Supervisor: I. M. Al Haris, S.Si, M.Si., II. Indah Manfaati Nur, S.Si, M.Si. Indonesia is an agricultural country that relies on the agricultural sector, especially rice plants. Demand for rice or unhulled rice fluctuated during the Covid-19 pandemic, causing the price of grain to experience instability. Changes in grain prices have linear data properties, apart from a linear pattern there is also a non-linear pattern so forecasting is carried out using the Hybrid ARIMAX-FFNN method with the Covid-19 dummy variable which has a significant influence on the harvested grain price data (GKP). From the result of the analysis, it is known that the average price of GKP in Indonesia for 173 months is around Rp. 4526,75.The best Hybrid ARIMAX-FFNN model on GKP price data in Indonesia is a combination of ARIMAX (0,1,0)(1,1,0)12 and residuals are modeled again with FFNN using the Backpropagation resilient algorithm.Modeling with FFNN obtained the best architecture,namely 12-13-1 (12 input neurons, 13 hidden neurons, 1 output neuron). The results of forecasting GKP prices in Indonesia in June 2022-May 2023 fluctuate every month. The lowest GKP price in Indonesia occurred in July 2022 at Rp.4310,93 and the hightest occurred in January 2023 at Rp. 5009,67. The prediction results of the Hybrid ARIMAX-FFNN model obtained a MAPE value of 1.75%. Keywords: Covid-19, Hybrid ARIMAX-FFNN, Price Of Harvested Dry Grain, Forecasting.

Actions (login required)

View Item