Optimising Trading Strategies using Linear Regression on Stock Prices

Stock price prediction is a challenging endeavour since it is highly dependent on stock demand, and there is no single characteristic that can correctly predict the demand of one stock each day. The proposed method intends to offer traders a reliable framework for optimizing their trading strategies based on historical data and linear regression analysis. This method involves identifying variables that affect stock prices, such as historical stock prices and economic indicators. The project outlines the steps to create two accurate models for predicting stock prices and EMA_10(Exponential Moving Average), which includes data cleaning, prioritization, training the model with appropriate algorithms like Ordinary Least Squares (OLS), and evaluating its performance using metrics like Root Mean Square Error (RMSE) and Mean Absolute Error(MAE). The proposed method is implemented using Python and can provide valuable insights into future trends of stock prices. With the proposed trading strategy investors can make informed decisions and minimize their risk in the stock market.