Accuracy of Moving Average Forecasting for NEPSE

Abstract

A simple moving average is one of the oldest and the simplest techniques of forecasting the trends of the stock market. The technical analysts follow mainly three types of moving averages, namely; simple, weighted, and exponential moving averages. Among these three types, as per the interest of investors, short-term and long-term time duration is used to calculate the trend using the moving average. All the mentioned moving averages are used by investors or analysts to predict the future trends of the market using historical data. Hence, for evaluating their forecasting accuracy, the paper has used both the short-term and the long-term moving average. The paper has used the NEPSE (closing) index values to calculate as well as plotted the moving averages to forecast the future trend and its accuracy with the help of Mean Absolute Percentage Error (MAPE). The paper found that there is a better crossover in the graphical representation of the moving average in the long-term moving average. In context to the Nepalese stock market, the MAPE results reflected a weekly (5-trading days) 5-SMA analysis of the market movement as the most relevant in short-term forecasting. Similarly, using the technique of moving average, 200-SMA (200-trading days of a year) was seen as the most effective to forecast long-term trends. The result of the long-term moving average MAPE pointed out that the annual reports of the listed companies better determine the trend of the market.