In recent years, the emergence of Machine Learning and Deep Learning has substantially improved the precision of stock price prediction, a critical area of interest for economists and investors. This research focuses on Nepal, a developing country with high economic volatility. We use Long Short-Term Memory (LSTM), a form of Recurrent Neural Network (RNN), and data from “sharesansar” to forecast stock prices. Our LSTM model has been modified to reduce losses while maintaining constant accuracy. This enhancement not only advantages experienced traders but also allows beginner traders to engage in lower-risk trading. We use popular metrics to assess the performance of our model, such as Root Mean Square Error (RMSE), Mean Square Error (MSE), and Mean Absolute Error (MAE). These measures are used to compare the accuracy of the LSTM model to classic techniques such as Support Vector Regression (SVR) and Auto-Regressive Integrated Moving Average (ARIMA). In the case of poor countries, our research shows that LSTM outperforms SVR and ARIMA models, giving greater accuracy with reduced error rates.