Forecasting the Search Trend of Muslim Clothing in Indonesia on Google Trends Data Using ARIMAX and Neural Network

Suhermi, N. and Suhartono, [Unknown] and Permata, R.P. and Rahayu, S.P. (2019) Forecasting the Search Trend of Muslim Clothing in Indonesia on Google Trends Data Using ARIMAX and Neural Network. In: Soft Computing in Data Science. Communications in Computer and Information Science . Springer, Singapore, pp. 272-286. ISBN 9789811503986

Abstract

The trend of muslim fashion has significantly raised the search trend for the brands of hijab and sarong in Indonesia. The aim of this study is to forecast the search trend for hijab and sarong based on google trends data. The Hijab brands include Rabbani, Zoya, Dian Pelangi, Elzatta, and Shafira, while the sarong brands include Gajah Duduk, Wadimor, Atlas, Mango, and Sapphire. We apply several forecasting methods such as Holt-Winters’ Exponential Smoothing, ARIMA, ARIMAX, FFNN and ERNN. The data contains calendar variation effect due to the Eid al-Fitr days use different calendar system. The results show that FFNN yields the most accurate forecast on 6 out of 10 brands. The forecast results for year 2019 period show that the search trend for Atlas brand is predicted to be the highest of all sarong brands. On the contrary, all the hijab brands’ trend search will decrease in this period.