Fuzzy Time Series Analysis for Stock Sales Forecasting

Investing is often considered one of the ways to generate profits by allocating funds to a place or company with the aim of gaining profits and avoiding inflation in the future. Among the many types of investments, investing in stocks is one of the popular ones. However, investing in the stock market is not easy because it is considered very risky due to the fluctuating prices of shares. The unstable movement of share prices is an important indicator for investors in determining whether they will sell, hold, or buy certain shares. Therefore, a method is needed to forecast the movement of share prices. This research aims to implement the fuzzy time series method for forecasting stock sales to support efficient decision-making. Using historical data on stock sales at PT. Bank Mandiri (Persero) from January 2, 2024, to October 4, 2024. The results of the study show that the application of the fuzzy time series method produces a forecast of stock price sales with a fairly high accuracy, with an accuracy of 98.7261%, and an MAPE error rate of only 1.2739% of the 180 data tested. This study shows that the forecasting model applied is able to provide an optimal picture of the relevant trends in the movement of stock sales, so it can be used to help make strategic decisions, thus it can be a reference for investors, especially in the stock field, to minimize risk in making decisions before investing.