Comparative Analysis of Stock Price Prediction using Time Series Models

The inconsistency of the stock market is ungovernable, making it prone to imminent financial risks. Our research indicates that the system displays remarkable precision in producing prognostics based on historical data derived from Yahoo Finance in the form of graphs and values. Resorting to our model, financial analysts and investors can alleviate risk and capitalize on market opportunities, facilitating informed decision-making. Our study’s primary focus is evaluating the following four models for predicting stock market prices: Long Short-Term Memory, Auto-Regressive Integrated Moving Average, Seasonal Auto-Regressive Integrated Moving Average using eXogenous factors, and Auto-Regressive Moving Average. ARIMA utilizes differencing to stabilize non-stationary time series, while ARMA amalgamates autoregressive and moving average components to record linear dependencies. SARIMAX strengthens ARIMA by integrating seasonal fluctuations, suitable for modeling seasonal trends in financial data. Deep learning techniques like LSTM excel at recognising complex temporal correlations. These findings showcase the importance of utilizing advanced forecasting techniques to navigate the ever-changing financial landscape, offering valuable insights for professional and academic purposes.