Forecasting PSEI using Holtwinter multiplicative model

Abstract

This study aims to determine and forecast stock prices in the Philippines using time series analysis, precisely the Holt-winter multiplicative method. The study proves that the HW-multiplicative model is better than the HW additive and simple moving average. The relative mean absolute error (MAE), mean square error (MSE) and mean absolute percentage error (MAPE) is the final indicators of the accuracy of forecasting models utilized. The margin of error between the Holt-winter Multiplicative, Holt-winter Additive, and Simple Moving Average models is compared to see which one is the most accurate. The lesser the error, the closer the stock market in the Philippines is to its actual trend. The data also demonstrated that the volatility of both price and non-price factors had a direct influence on the movement of the Philippine stock market. By displaying graphical patterns and thorough tables of data, this study ensures that predicting is free of bias.