Modified Weighted for Enrollment Forecasting Based on Fuzzy Time Series

Abstract

The enrollment study is main point for the university planning. Many previous studies have been presented for enrollment forecasting. This paper proposed the adoption the weighted and the difference between actual data toward midpoint interval based on fuzzy time series. By using the enrollment of the University of Alabama and Universiti Teknologi Malaysia (UTM), as data sets for training and testing, the performance of the adoption approach has been shown much improvement in terms of MSE (Mean Square Error) and average error of forecasting measurements. Keywords: Recurrence; chronological order; modified weighted; enrollment; fuzzy time series.