On Optimization and Analysis of The Moving Average Crossover Trading Strategy in the Nepal Stock Exchange (NEPSE)

This study investigates the effectiveness of various Moving Averages (MA) trading algorithms, namely Simple Moving Average (SMA), Exponential Moving Average (EMA), and Triple Smoothed Exponential Moving Average (TSEMA), in the context of the Nepal Stock Exchange (NEPSE). Stocks were classified into three categories: High Cap, Mid Cap, and Low Cap, based on market capitalization. The optimal short- and long-term window parameters (n1, n2) were determined using historical price data for each category. The algorithms were evaluated under both ideal and real-world conditions, the latter incorporating transaction costs and varying levels of investment capital (Rs.50,000 and Rs.1,00,00,000). Performance was assessed using multiple metrics, including average profit percentage, Sharpe ratio, maximum drawdown, and the average number of trades executed. Results indicate that TSEMA consistently outperforms SMA and EMA across most categories, offering superior returns and stability while maintaining lower trading frequency. Although the Buy & Hold strategy shows higher absolute returns in certain cases, MA-based strategies demonstrate improved risk-adjusted performance, particularly in volatile market segments. These findings suggest that adaptive MA-based trading frameworks such as TSEMA can provide viable, data-driven alternatives to passive investment strategies in emerging markets like Nepal