The Nasdaq Composite Index Prediction Using LSTM and Bi-LSTM Multivariate Deep Learning Approaches

The market values of over 2,500 stocks that are traded on the Nasdaq stock exchange are combined to form the Nasdaq Composite Index, which uses a market capitalization-weighted system and runs under a market value-weighted system. Notably, the individual market value of each stock is factored into the calculation of this index. Companies having headquarters both in the United States and elsewhere in the world are included in the index's coverage. In spite of the fact that the technology sector has a significant impact on the index, the credibility of the index rests on its extensive representation across a variety of industries. Investigating the predictive skills of the Nasdaq Composite Index is absolutely necessary in light of the significant influence that its performance has on the economy of the entire world. Through the utilisation of a time series-based predictive analysis, this piece dives into the topic of forecasting the performance of an index that is traded on a national stock market. Utilising a real-time web data source allowed for the acquisition of the most recent data available up until the year 2002. Within the scope of this study, multivariate algorithms, more especially LSTM and Bidirectional-LSTM, were utilised to investigate and analyse prospective patterns and forecasts relating to the cost of petrol. The research unequivocally shows that, when it comes to forecasting stock prices, the Bidirectional-LSTM model is superior to the LSTM model. This is demonstrated by the analysis.