Claim Missing Document
Check
Articles

Found 2 Documents
Search

Application of ARIMA to Curly Red Chili Prices in Bengkulu City Melda Juliza; Puce Angreni
INSOLOGI: Jurnal Sains dan Teknologi Vol. 2 No. 2 (2023): April 2023
Publisher : Yayasan Literasi Sains Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55123/insologi.v2i2.1871

Abstract

Curly red chili in Bengkulu City often experiences price fluctuations from time to time. These price fluctuations are sometimes very extreme, causing public unrest both for food processing industry entrepreneurs and for daily household needs. Therefore, this study uses time series techniques to predict the price of curly red chili in Bengkulu City. This study discusses chili price forecasting using the Box-Jenkins ARIMA model based on curly red chili price data in Bengkulu City from 03 October 2022 to 28 February 2023. This research aims to look at the accuracy of the best model for curly red chili prices in Bengkulu city for the ARIMA model based on ACF & PACF criteria with autocorrelation coefficient values, and the smallest AIC criteria with the auto.arima function in R software. Next, forecast the price of curly red chili in Bengkulu City for the next period with the ARIMA model based on the best criteria obtained. Based on the ADF test, it can be seen that the data is not stationary so the data differencing process is carried out. The analysis results show that the best ARIMA model for curly red chili price data in Bengkulu City is the automatic ARIMA model with the smallest AIC criteria using the auto.arima function with the value of RMSE is 4197.7. The ARIMA model that is formed is the ARIMA (1,1,1) model. Next, the results of forecasting the price of curly red chili for Bengkulu City obtained based on the ARIMA (1,1,1) on 01 March 2023 is Rp 41,700.
Comparison of Methods ARIMA and MAR Models with MODWT Decomposition on Non-Stationary Data Angreni, Puce; Melda Juliza
INSOLOGI: Jurnal Sains dan Teknologi Vol. 2 No. 2 (2023): April 2023
Publisher : Yayasan Literasi Sains Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55123/insologi.v2i2.1888

Abstract

The forecasting methods used in this study are Autoregressive Integrated Moving Average (ARIMA) and Multiscale Autoregressive (MAR). The ARIMA model does not include predictor variables in the model. The MAR model is a model that performs the transformation process using wavelets. The MAR model adopts an autoregressive time series (AR) model with wavelet coefficients and scale coefficients as predictors. The wavelet coefficient and scale are obtained by decomposition using Maximal Overlap Discrete Wavelet Transformation (MODWT). MODWT functions to describe data based on the level of each wavelet filter. This study aims to determine the best forecasting model using ARIMA and MAR models. The time series data used in this study is data on the rupiah exchange rate against the US dollar. Data on the rupiah exchange rate against the US Dollar for 2019-2020 is non-stationary data, so the ARIMA and MAR models can be used in this study.