Forecasting Search Trends for “Silverqueen” Chocolate Keywords using the Singular Spectrum Analysis Method and the Hybrid Singular Spectrum Analysis-ARIMA Model
Article Sidebar
Main Article Content
Abstract
This research uses a hybrid model of time series in the process. The Singular Spectrum Analysis (SSA) method will be combined with the Autoregressive Integrated Moving Average (ARIMA) method to model noise from SSA. This research applied the hybrid SSA-ARIMA method to Google Trends data, especially to the chocolate keyword "Silverqueen" search trend. This research aims to assess the accuracy and identify the best forecasting method for search trends for the keyword "Silverqueen" chocolate in Indonesia. Based on the results, the accuracy value obtained for the SSA method was 0.54% (MAPE) and 0.04 (RMSE) for in-sample data and 28.93% (MAPE) and 1.49 (RMSE) for out-sample data. The hybrid SSA-ARIMA (5.1.0) method has two outliers with an accuracy value of 0.35% (MAPE) and 0.02 (RMSE) for insample data and 31.00% (MAPE) and 1.50 (RMSE) for out-sample data. The results of the SSA forecasting method for the next 17 periods show that the trend will increase, with the highest trend occurring in the second week of February 2024, namely 100 points. Then, the forecast results of the hybrid SSA-ARIMA(5,1,0) method with outliers for the next 17 periods, the trend will increase, with the highest trend occurring in the second week of February 2024, namely around 95 points. The best method for forecasting search trends for the chocolate keyword “Silverqueen” is the SSA method.
Article Details
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
References
Goyaldina N, Nekrutkin V, Zhigljavsky A. Analysis of Time Series Structure: SSA and Related Techniques. United States of America: Chapman&Hall/CRC; 2001.
Ischak R, Asrof A, Darmawan G. Peramalan Rata-Rata Harga Beras di Tingkat Penggilingan Menggunakan Model Singular Spectrum Analysis (SSA). Prosiding Seminar Nasional Matematika dan Pendidikan Matematika. Yogyakarta; 2018.
Puspita W, Rustiana S, Suparman, Purwandari T. Perbandingan Hasil Peramalan Curah Hujan Bulanan Kota Bogor Dengan Seasonal Autoregressive Integrated Moving Average (SARIMA) dan Singular Spectrum Analysis (SSA). Prosiding SENDIKA. 2019;5(2):206-17.
Hidayat K, Wahyuningsih S, Nasution Y. Pemodelan Jumlah Titik Panas di Provinsi Kalimantan Timur Dengan Metode Singular Spectrum Analysis. Jambura Journal of Probability and Statistics. 2020;1(2):78-88.
Lai Y, Dzombak D, Use of the Autoregressive Integrated Moving Average (ARIMA) Model to Forecast Near-Term Regional Temperature and Precipitation. American Meteorological Society Journals. 2020;35(3):959-76.
Kusumaningrum N, Purnamasari I, Siringoringo M. Peramalan Menggunakan Model Hybrid ARIMAX-NN untuk Total Transaksi Pembayaran Nontunai. Journal Of Statistics and Its Application on Teaching and Research. 2023;5(1):1-14.
Ilahi E, Zukhronah E, Susanti Y. Model Hibrida Singular Spectrum Analysis (SSA) dan Autoregressive Integrated Moving Average (ARIMA) untuk Peramalan Indeks Harga Konsumen. Jurnal Seminar Nasional Pendidikan Matematika Ahmad Dahlan. 2023;7(11):72-81.
Kumar U. An Integrated SSA-ARIMA Approach to Make Multiple Day Ahead Forecast for the Daily Maximum Ambient O3 Concentration. Aerosol and Air Quality Research. 2015;15(1): 208-19.
Arumsari M, Wahyuningsih S, Siringoringo M. Peramalan Inflasi Provinsi Kalimantan Timur Menggunakan Model Hybrid Singular Spectrum Analysis Autoregressive Integrated Moving Average. Jurnal Matematika Sains dan Komputasi. 2021;18(1):78-92.
Chumnumpan P, Shi X, Understanding New Products’ Market Performance Using Google Trends. Australian Marketing Journal. 2019;27(2):91-103.
Boone T, Ganeshan R, Jain A, Sanders N. Forecasting Sales In The Supply Chain: Consumer Analytics In The Big Data Era. International Journal of Forecasting. 2019;35(1):170-80.
Diksa I. Forecasting the Existence of Chocolate with Variation and Seasonal Calendar Effects Using the Classic Time Series Approach. Jurnal Matematika, Statistika & Komputasi. 2022;18(2):237-50.
Fitri A, Anindita R, Nugroho C. Analisis Tingkat Kepuasan Konsumen dan Loyalitas Merek Cokelat Silverqueen di Kabupaten Pekalongan. Jurnal Ekonomi Pertanian dan Agribisnis. 2023;7(1):127-45.
Husnita F, Wahyuningsih S, Nohe D. Analisis Spektral dan Model ARIMA untuk Peramalan Jumlah Wisatawan di Dunia Fantasi Taman Impian Jaya Ancol. Jurnal Eksponensial. 2015;6(1): 21-9.
Wei W. Time Series Analysis: Univariate and Multivariate Methods. New York: Pearson Education; 2006.
Khaeri H, Yulian E, Darmawan G. Penerapan Metode Singular Spectrum Analysis (SSA) Pada Peramalan Jumlah Penumpang Kereta Api Di Indonesia Tahun 2017. Jurnal Euclid. 2018;5(1):8-20.
Wulandari G, Wahyuningsih S, Siringoringo M, Sergio A. Inflation Forecasting for Samarinda City Using Hybrid Singular Spectrum Analysis-Neural Network Model. The 4th International Conference on Mathematics and Sciences: AIP Conferences Proceedings. 2024;3095(1).
Hassani H, Mohamoudvand R. Singular Spectrum Analysis With R. Iran: Palgrave Advanced Texts in Econometrics; 2018.
Aswi , Sukarna. Analisis Runtun Waktu dan Teori. Makassar: Andira Publisher; 2006.
Suparti, Sa'adah A. Analisis Data Inflasi Indonesia Menggunakan Model Autoragressive Integrated Moving Average (ARIMA) Dengan Penambahan Outlier. Media Statistika. 2015;8(1):1-11.
Waeto S, Chuarkham K, Intarasit A. Forecasting Time Series Movement Direction with Hybrid Methodology. Journal of Probability and Statistics. 2017; 2-3.