METODE FUZZY TIME SERIES MUSIMAN BERDASARKAN PARTISI INTERVAL FREKUENSI DENSITAS

(A Seasonal Fuzzy Time Series Based on Frequency Density Interval Partitioning)

Abstract

Time series data can be used as material to predict the probability of future events. Time series data has several patterns, one of which is a seasonal pattern. In processing time series data, an analytical method is needed. The fuzzy time series method can be used to analyze time series data using the concept of fuzzy logic. Some fuzzy time series methods usually produce large errors if the data being tested has a seasonal pattern. Therefore, a seasonal fuzzy time series method was developed that can be used for time series data with seasonal patterns. In the fuzzy time series method, it is necessary to determine the effective interval length in order to obtain optimal accuracy. In this study, the frequency density interval partitioning was used to determine the length of the interval. The purpose of this study is to examine the seasonal fuzzy time series method based on the frequency density interval partition. The results of this study indicate that the seasonal fuzzy time series method is suitable for processing seasonal patterned data and the determination of interval length using frequency density interval partitioning can provide optimal accuracy. 

Keywords: Frequency Density, Fuzzy Time Series, Interval Partition, Seasonal Fuzzy Time Series, Time Series