Application of Fuzzy Time Series Chen and Cheng Methods to Forecast Profit in a State-Owned Insurance Company

PT. Taspen (Persero) as a state-owned enterprise in the services sector needs to analyze financial performance to understand the fluctuations in quarterly profit in 2022. This study uses the Fuzzy Time Series (FTS) forecasting method with two approaches, namely Fuzzy Time Series Chen and Fuzz Time Series Cheng, to predict profit and dividend prospects. The analysis stage of Chen's Fuzzy Time Series method includes determining the set of the universe (U), forming intervals, defining fuzzy sets, determining the membership value of each data, fuzzification of data, formation of Fuzzy Logic Relationships (FLR) and Fuzzy Logic Relationship Groups (FLRG), forecasting and defuzzification. Meanwhile, Cheng's Fuzzy Times Series method has similar stages but is equipped with FLRG weighting into the W matrix and standardization of the W matrix*. The results of the calculation of forecasting accuracy through MAPE, MSE, and MAE show that Cheng's Fuzzy Time Series method is more accurate than Chen's Fuzzy Time Series, with a smaller error value. This confirms that Cheng's Fuzzy Times Series method is more reliable in projecting PT. Taspen (Persero).