Data Analysis, Visualization and Prediction of Stock Market Prices of K-Electric

Abstract

Predicting stock price is a trend yet very challenging task. It is because the stock prices depend upon several internal and external factors. Stock price prediction can be very useful for financial sectors and the government and help in informed decision-making. This paper analyzes the stock market prices of K-Electric Karachi. It is found that the stock prices of K-electric depend on the stock prices of the refinery sector. The paper analyzes the stock price data of the two sectors. Also, the paper compares the stock price prediction based on moving average, auto-regressive integrated moving average (ARIMA), convolutional neural network and long short-term memory (LSTM) model. It is found that ARIMA outperforms the other algorithms. A set of experiments were conducted to test the performance of algorithms. The algorithms were analyzed based on different metrics such as root mean square error (RMSE), mean absolute error (MAE) and mean absolute percentage error (MAPE).