Average-based intervals and frequency density-based intervals in forecasting tuberculosis cases in Sabah

• Main Authors: Suriana Lasaraiya, Suzelawati Zenian, Risman Mat Hasim
• Format: Proceedings
• Language: English; English
• Published: Pusat e-pembelajaran, UMS 2021
• Subjects: Q1-295 General; RC306-320.5 Tuberculosis
• Online Access: https://eprints.ums.edu.my/id/eprint/41613/1/ABSTRACT.pdf; https://eprints.ums.edu.my/id/eprint/41613/2/FULL%20TEXT.pdf; https://eprints.ums.edu.my/id/eprint/41613/; https://oer.ums.edu.my/handle/oer_source_files/1874
• Institution: Universiti Malaysia Sabah

Initially, time series models were used to forecast the number of students enrolled in the University of Alabama in 1993. Forecasting is one of the branches of fuzzy sets theory. As time goes on, these models are being used to make predictions of stock prices, weather, road accidents, and several other models. In this paper, we compare two different approaches in determining the suitable length of intervals to increase the accuracy of forecasting in fuzzy time series. The methods proposed are the average-based intervals and frequency-density-based partitioning. The results showed that the average-based intervals have higher accuracy in forecasting the number of cases compared to frequency-density-based intervals.