High stock price fluctuations pose a significant challenge for investors and analysts in determining investment strategies. Price dynamics influenced by economic, political, and psychological market factors require forecasting methods that can accommodate uncertainty and non-linear historical data patterns. This study applies Cheng's Fuzzy Time Series algorithm to predict the stock price of PT. ABCD by going through the stages of universe set formation, interval determination, fuzzification, fuzzy logic relationship formation, and defuzzification to obtain prediction results. The method implementation was carried out using two approaches: manual calculation using Microsoft Excel and automatic calculation using the Orange application. The results show that Cheng's method is able to produce predictions very close to the actual value, with an accuracy level measured using the Mean Absolute Percentage Error (MAPE) indicator of 0.058787% on both platforms. The consistency of the results between Excel and Orange proves the reliability of Cheng's method, so it can be used as a reference in supporting investment decision-making in the Indonesian capital market.
Copyrights © 2026