A Study of Time Series Forecasting Enrollments Using Fuzzy Interval Partitioning Method

Abstract

A time series is a sequence of elements with numerical data in sequential order and having regular intervals. Time series are used in statistics, enrollments, signal processing, econometrics, mathematical finance and weather forecasting, etc. It helpsus to forecast and predict the time series data in different domains. There are many methods to forecast the enrollments in literature which have large applications and are presented in the field of statistics and econometrics. One of the robust methods, we used in our research is moving average. It helps to forecast and predict the data whenever the fuzziness occurs in time series data, which is not appropriate in crisp time series forecasting. To get rid of this problem, the fuzzy interval partitioning method proved to be more appropriate to generate accurate results. This research will focus to overcomethe failure of the crisp method and to show the use of a fuzzy interval partitioning method. The fuzzy interval partitioning methodis different from another interval partition schemes because it specifies the linguistic values rather than numerical values. It is also used todeal with uncertain conditions. So, fuzzy interval partitioning improves data utilization and also calculate the higher predicted accuracy rate. Besides this research, we use a quantitative method and a fuzzy moving average with the interval partitioning method. Then we compare the efficiency of moving average model and moving average with fuzzy interval partitioning method for forecasting the enrollments.

Received: 15 January 2022 | Revised: 8 March 2022 | Accepted: 16 March 2022

Conflicts of Interest

The authors declare that they have no conflicts of interest to this work.