Analysis and forecasting of Time-Series data using S-ARIMA, CNN and LSTM

Analyzing the behavior of stock market movements has often been an area of interest to machine learning and time-series data analyst. It has been very challenging due to its immense complex nature, chaotic, and dynamic environment. With the advent of machine learning and deep learning algorithms, this paper aims to significantly reduce the risk of trend prediction. This study compares models for Time - Series forecasting i.e. SARIMA (Seasonal Auto-Regressive Integrated Moving Average), CNN (Convolutional Neural Network), and LSTM (Long Short-Term Memory) for predicting Nifty-500 indices trend. The results that were obtained are promising and the evaluation unveils the power of Deep Learning through CNN and LSTM but also empowers the S-ARIMA model, making a great impact on the Machine Learning paradigm.