MODEL SEASONAL AUTOREGRESSIVE INTEGRATED MOVING AVERAGE (SARIMA) PADA PERAMALAN METODE FUZZY TIME SERIES MARKOV CHAIN (FTS-MC)

GINDA ATI SUWANDI, 1717031064 (2021) MODEL SEASONAL AUTOREGRESSIVE INTEGRATED MOVING AVERAGE (SARIMA) PADA PERAMALAN METODE FUZZY TIME SERIES MARKOV CHAIN (FTS-MC). FAKULTAS MATEMATIKA ILMU PENGETAHUAN ALAM, UNIVERSITAS LAMPUNG.

[img]
Preview
FIle PDF
ABSTRAK-ABSTRACT - GINDA.pdf

Download (4Mb) | Preview
[img] FIle PDF
SKRIPSI FULL - GINDA.pdf
Restricted to Hanya staf

Download (4Mb)
[img]
Preview
FIle PDF
SKRIPSI TANPA BAB PEMBAHASAN - GINDA.pdf

Download (4Mb) | Preview

Abstrak

Metode SARIMA merupakan pengembangan dari metode Box-Jenkins (ARIMA). Model SARIMA dapat mengatasi pola musiman dari suatu periode waktu. Model ini memerlukan beberapa pendekatan seperti asumsi kestasioneran, pembedaan (differencing), dan transformasi data. Namun, pendekatan ini masih belum mampu mengurangi nilai kesalahan model, akibatnya akan mendapatkan hasil peramalan dengan error yang besar. Pada proses peramalan model FTS-MC terdapat perhitungan nilai penyesuaian yang bertujuan untuk mengurangi besarnya penyimpangan hasil peramalan. Penelitian ini bertujuan untuk mengetahui apakah metode fuzzy time series Markov chain dapat memperbaiki hasil peramalan model SARIMA. Berdasarkan hasil penelitian diperoleh model ARIMA(0,1,1)(1,1,1)12 sebagai model terbaik yang akan digunakan untuk peramalan bulan Januari 2015 – Juli 2016. Hasil peramalan bulan Juli 2016 akan digunakan sebagai proses peramalan FTS-MC. Nilai MAPE yang diperoleh dari kedua model sama-sama dibawah 10% yang berarti hasil peramalan sangat baik. Namun, nilai MAPE dari metode FTS-MC lebih kecil dibandingkan model SARIMA. Hal ini menunjukkan bahwa metode FTS-MC dapat memperbaiki hasil peramalan model SARIMA. Kata kunci: fuzzy time series, fuzzy time series markov chain, SARIMA The SARIMA method is a development of the Box-Jenkins (ARIMA) method. The SARIMA model can overcome the seasonal pattern at a period of time. This model requires several approaches such as stationarity assumption, differencing, and data transformation. However, this approach is still not able to reduce the error value of the model, as a result it will get forecasting results with errors large. In the process of forecasting the FTS-MC model there is a calculation of the adjustment value which aims to reduce the magnitude of the deviation of the forecasting results. This study aims to determine whether the method fuzzy time series Markov chain can improve the forecasting results of the SARIMA model. Based on the research results, ARIMA(0,1,1)(1,1,1)12 model is the best model to be used for forecasting January 2015 – July 2016. The forecasting results for July 2016 will be used as the FTS-MC forecasting process. The MAPE values obtained from both models are both below 10%, which means the forecasting results are very good. However, the MAPE value of the FTS-MC method is smaller than the SARIMA model. This shows that the FTS-MC method can improve the forecasting results of the SARIMA model. Keywords: fuzzy time series, fuzzy time series markov chain, SARIMA

Jenis Karya Akhir: Skripsi
Subyek: 500 ilmu pengetahuan alam dan matematika
Program Studi: Fakultas MIPA > Prodi Matematika
Pengguna Deposit: UPT . Neti Yuliawati
Date Deposited: 24 May 2022 06:39
Terakhir diubah: 24 May 2022 06:39
URI: http://digilib.unila.ac.id/id/eprint/61743

Actions (login required)

Lihat Karya Akhir Lihat Karya Akhir