Fuzzy Time Series Application in Predicting the Number of Confirmation Cases of Covid-19 Patients in Indonesia

Lintang Patria (Universitas Terbuka)

International Journal of Quantitative Research and Modeling
Vol 2, No 4 (2021)

Forecasting is a statistical method that can use historical data patterns to predict future events. This article discusses the prediction of the number of new confirmed cases of Covid-19 patients in Indonesia. The data used is from January 1, 2021 to August 7, 2021. The methods used are Fuzzy Time Series (FTS) Chen (2014) and Cheng et al. (2008). FTS is a forecasting method that uses rules and logic on fuzzy sets. The level of prediction accuracy is then calculated based on the Mean Absolute Percentage Error (MAPE) value. The MAPE values of these two methods are then compared to know which method is more suitable in this case study. The results showed that FTS Chen produced an accuracy of 12.75% and FTS Cheng produced an accuracy of 14.27%. The results of this study indicate that FTS Chen and FTS Cheng produce good accuracy and can be used to predict new confirmed cases of Covid 19 sufferers in Indonesia.