COMPARATIVE ANALYSIS OF TIME SERIES FORECASTING MODELS USING ARIMA AND NEURAL NETWORK AUTOREGRESSION METHODS
Gold price fluctuations have a significant impact because gold is a haven asset. When financial markets are volatile, investors tend to turn to safer instruments such as gold, so gold price forecasting becomes important in economic uncertainty. The novelty of this research is the comparative analysi...
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Universitas Pattimura
2024-10-01
|
| Series: | Barekeng |
| Subjects: | |
| Online Access: | https://ojs3.unpatti.ac.id/index.php/barekeng/article/view/13128 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849770452901167104 |
|---|---|
| author | Melina Melina Sukono Sukono Herlina Napitupulu Norizan Mohamed Yulison Herry Chrisnanto Asep ID Hadiana Valentina Adimurti Kusumaningtyas Ulya Nabilla |
| author_facet | Melina Melina Sukono Sukono Herlina Napitupulu Norizan Mohamed Yulison Herry Chrisnanto Asep ID Hadiana Valentina Adimurti Kusumaningtyas Ulya Nabilla |
| author_sort | Melina Melina |
| collection | DOAJ |
| description | Gold price fluctuations have a significant impact because gold is a haven asset. When financial markets are volatile, investors tend to turn to safer instruments such as gold, so gold price forecasting becomes important in economic uncertainty. The novelty of this research is the comparative analysis of time series forecasting models using ARIMA and the NNAR methods to predict gold price movements specifically applied to gold price data with non-stationary and non-linear characteristics. The aim is to identify the strengths and limitations of ARIMA and NNAR on such data. ARIMA can only be applied to time series data that are already stationary or have been converted to stationary form through differentiation. However, ARIMA may struggle to capture complex non-linear patterns in non-stationary data. Instead, NNAR can handle non-stationary data more effectively by modeling the complex non-linear relationships between input and output variables. In the NNAR model, the lag values of the time series are used as input variables for the neural network. The dataset used is the closing price of gold with 1449 periods from January 2, 2018, to October 5, 2023. The augmented Dickey-Fuller test dataset obtained a p-value = 0.6746, meaning the data is not stationary. The ARIMA(1, 1, 1) model was selected as the gold price forecasting model and outperformed other candidate ARIMA models based on parameter identification and model diagnosis tests. Model performance is evaluated based on the RMSE and MAE values. In this study, the ARIMA(1, 1, 1) model obtained RMSE = 16.20431 and MAE = 11.13958. The NNAR(1, 10) model produces RMSE = 16.10002 and MAE = 11.09360. Based on the RMSE and MAE values, the NNAR(1, 10) model produces better accuracy than the ARIMA(1, 1, 1) model. |
| format | Article |
| id | doaj-art-a5e8a906256644fb8a5e463bdc8be0a1 |
| institution | DOAJ |
| issn | 1978-7227 2615-3017 |
| language | English |
| publishDate | 2024-10-01 |
| publisher | Universitas Pattimura |
| record_format | Article |
| series | Barekeng |
| spelling | doaj-art-a5e8a906256644fb8a5e463bdc8be0a12025-08-20T03:02:59ZengUniversitas PattimuraBarekeng1978-72272615-30172024-10-011842563257610.30598/barekengvol18iss4pp2563-257613128COMPARATIVE ANALYSIS OF TIME SERIES FORECASTING MODELS USING ARIMA AND NEURAL NETWORK AUTOREGRESSION METHODSMelina Melina0Sukono Sukono1Herlina Napitupulu2Norizan Mohamed3Yulison Herry Chrisnanto4Asep ID Hadiana5Valentina Adimurti Kusumaningtyas6Ulya Nabilla7Department of Informatics, Faculty of Science and Informatics, Universitas Jenderal Achmad Yani, IndonesiaDepartment of Mathematics, Faculty of Mathematics and Natural Sciences, Universitas Padjadjaran, IndonesiaDepartment of Mathematics, Faculty of Mathematics and Natural Sciences, Universitas Padjadjaran, IndonesiaDepartment of Mathematics, Faculty of Ocean Engineering Technology and Informatics, Universiti Malaysia Terengganu, MalaysiaDepartment of Informatics, Faculty of Science and Informatics, Universitas Jenderal Achmad Yani, IndonesiaDepartment of Informatics, Faculty of Science and Informatics, Universitas Jenderal Achmad Yani, IndonesiaDepartment of Chemistry, Faculty of Science and Informatics, Universitas Jenderal Achmad Yani, IndonesiaDepartment of Mathematics, Faculty of Engineering, Universitas Samudra, IndonesiaGold price fluctuations have a significant impact because gold is a haven asset. When financial markets are volatile, investors tend to turn to safer instruments such as gold, so gold price forecasting becomes important in economic uncertainty. The novelty of this research is the comparative analysis of time series forecasting models using ARIMA and the NNAR methods to predict gold price movements specifically applied to gold price data with non-stationary and non-linear characteristics. The aim is to identify the strengths and limitations of ARIMA and NNAR on such data. ARIMA can only be applied to time series data that are already stationary or have been converted to stationary form through differentiation. However, ARIMA may struggle to capture complex non-linear patterns in non-stationary data. Instead, NNAR can handle non-stationary data more effectively by modeling the complex non-linear relationships between input and output variables. In the NNAR model, the lag values of the time series are used as input variables for the neural network. The dataset used is the closing price of gold with 1449 periods from January 2, 2018, to October 5, 2023. The augmented Dickey-Fuller test dataset obtained a p-value = 0.6746, meaning the data is not stationary. The ARIMA(1, 1, 1) model was selected as the gold price forecasting model and outperformed other candidate ARIMA models based on parameter identification and model diagnosis tests. Model performance is evaluated based on the RMSE and MAE values. In this study, the ARIMA(1, 1, 1) model obtained RMSE = 16.20431 and MAE = 11.13958. The NNAR(1, 10) model produces RMSE = 16.10002 and MAE = 11.09360. Based on the RMSE and MAE values, the NNAR(1, 10) model produces better accuracy than the ARIMA(1, 1, 1) model.https://ojs3.unpatti.ac.id/index.php/barekeng/article/view/13128arimaforecastingnnarnon-lineartime series |
| spellingShingle | Melina Melina Sukono Sukono Herlina Napitupulu Norizan Mohamed Yulison Herry Chrisnanto Asep ID Hadiana Valentina Adimurti Kusumaningtyas Ulya Nabilla COMPARATIVE ANALYSIS OF TIME SERIES FORECASTING MODELS USING ARIMA AND NEURAL NETWORK AUTOREGRESSION METHODS Barekeng arima forecasting nnar non-linear time series |
| title | COMPARATIVE ANALYSIS OF TIME SERIES FORECASTING MODELS USING ARIMA AND NEURAL NETWORK AUTOREGRESSION METHODS |
| title_full | COMPARATIVE ANALYSIS OF TIME SERIES FORECASTING MODELS USING ARIMA AND NEURAL NETWORK AUTOREGRESSION METHODS |
| title_fullStr | COMPARATIVE ANALYSIS OF TIME SERIES FORECASTING MODELS USING ARIMA AND NEURAL NETWORK AUTOREGRESSION METHODS |
| title_full_unstemmed | COMPARATIVE ANALYSIS OF TIME SERIES FORECASTING MODELS USING ARIMA AND NEURAL NETWORK AUTOREGRESSION METHODS |
| title_short | COMPARATIVE ANALYSIS OF TIME SERIES FORECASTING MODELS USING ARIMA AND NEURAL NETWORK AUTOREGRESSION METHODS |
| title_sort | comparative analysis of time series forecasting models using arima and neural network autoregression methods |
| topic | arima forecasting nnar non-linear time series |
| url | https://ojs3.unpatti.ac.id/index.php/barekeng/article/view/13128 |
| work_keys_str_mv | AT melinamelina comparativeanalysisoftimeseriesforecastingmodelsusingarimaandneuralnetworkautoregressionmethods AT sukonosukono comparativeanalysisoftimeseriesforecastingmodelsusingarimaandneuralnetworkautoregressionmethods AT herlinanapitupulu comparativeanalysisoftimeseriesforecastingmodelsusingarimaandneuralnetworkautoregressionmethods AT norizanmohamed comparativeanalysisoftimeseriesforecastingmodelsusingarimaandneuralnetworkautoregressionmethods AT yulisonherrychrisnanto comparativeanalysisoftimeseriesforecastingmodelsusingarimaandneuralnetworkautoregressionmethods AT asepidhadiana comparativeanalysisoftimeseriesforecastingmodelsusingarimaandneuralnetworkautoregressionmethods AT valentinaadimurtikusumaningtyas comparativeanalysisoftimeseriesforecastingmodelsusingarimaandneuralnetworkautoregressionmethods AT ulyanabilla comparativeanalysisoftimeseriesforecastingmodelsusingarimaandneuralnetworkautoregressionmethods |