Perbandingan Metode-metode Peramalan Statistika untuk Data Indeks Harga Pangan

Aris Gunaryati; Fauziah Fauziah; Septi Andryana

doi:10.30998/string.v2i3.2200

Perbandingan Metode-metode Peramalan Statistika untuk Data Indeks Harga Pangan

Aris Gunaryati^(1*), Fauziah Fauziah⁽²⁾, Septi Andryana⁽³⁾

(1) Fakultas Teknologi Komunikasi dan Informatika - Universitas Nasional
(2) Fakultas Teknologi Komunikasi dan Informatika - Universitas Nasional
(3) Fakultas Teknologi Komunikasi dan Informatika - Universitas Nasional
(*) Corresponding Author

Abstract

The food issue is very important to predict the price of food, especially rice, in the future. This is because rice is the Indonesian staple food, the price of which must be always made stable and affordable for the community. Therefore, it is necessary to make a proper forecasting model to know the approximate price of rice in the future. The forecasting method that is often used in researches is a quantitative forecasting method with time series data. This research is conducted to make comparison among the statistic forecasting methods, namely trend analysis, exponential smoothing and decomposition to analyze the development of average rice price at Indonesian wholesaler level from 2010 to 2016. The existing data shows that the most suitable forecasting model for the average data of rice price at Indonesian wholesaler level in 2010 to 2016 is Double Exponential Smoothing (Brown), with MSE level of 188,086.086. and that the prediction about average rice price in 2017 reaches Rp. 11600,- with the actual price of Rp. 11534,-

Keywords

decomposition, exponential smoothing, forecasting, MSE, trend

Full Text:

PDF (Indonesian)

References

Lim Sanny, Metode Forecasting dengan Pendekatan Program Minitab, Prosiding Seminar Nasional dan Call for Paper Pengentasan Kemiskinan Melalui UMKM : Komparasi Model Indonesia dan Malaysia, UPN Veteran Yogyakarta, 2012.

DT Wiyanti, R Pulungan, Peramalan Deret Waktu Menggunakan Model Fungsi Basis Radial (RBF) dan Auto Regessive Integrated Moving Average (ARIMA), Jurnal MIPA 35 (2) : 175-182. 2012.

Zhang X, Liu Y, Yang M, Zhang T, Young AA, Li X. Comparative Study of Four Time Series Methods in Forecasting Typhoid Fever Incidence in China. Vespignani A, ed. PLoS ONE. 2013; 8(5) : e63116. doi:10.1371/journal.pone.0063116.

Atok, R.M. and Suhartono. Comparison between Neural Networks, ARIMA Box-Jenkins and Exponential Smoothing Methods for Time Series Forecasting. Research Report, Lemlit: Sepuluh Nopember Institute of Technology. 2000.

Suhartono, Subanar and Guritno, S. (2005a). A Comparative Study of Forecasting Models for Trend and Seasonal Time Series: Does complex model always yield better forecast than simple models? JURNAL TEKNIK INDUSTRI: Jurnal Keilmuan dan Aplikasi Teknik Industri, Vol. 7, No. 1, pp. 27-41. 2005.

Ayodele Ariyo Adebiyi, dkk., “Comparison of ARIMA and Artificial Neural Networks Models for Stock Price Prediction ‘’, Journal of Applied Mathematics, vol. 2014, Article ID 614342, 7 pages, 2014. doi:10.1155/2014/614342

DOI: http://dx.doi.org/10.30998/string.v2i3.2200