SPIC: a stock price indicator based on crises prediction using bi-directional LSTM

The causes and consequences of stock market collapses have been the subject of several researches. Some of the explanations might be attributed to economic factors such as “rising interest rates”, “high inflation”, or a “recession”. Political uncertainty, natural disasters, or a crisis in a specific industry might potentially be the cause. These studies primarily concentrate on stock price prediction across all main indexes such as business profits, geopolitical unrest, the financial crisis, and pandemic conditions. The process of predicting a stock crisis is challenging since the stock market is more volatile than usual. The challenge of crisis prediction is difficult for academics and investors. The price history and trend volume of 50 stocks have been collected from “The national stock exchange” of India. First, the irrelevant financial parameters are removed using “principal component analysis”. The second is the bi-directional LSTM deep neural for regression approach, which is used to categorize stocks with solid fundamentals. The third is the identification and detection of bubble and stock crisis events using the “moving average crossover”. The fourth method uses a bi-directional LSTM deep neural “convolutional neural network” and “Bayesian linear regression” technique to anticipate stock crises. “mean squared error”, “mean absolute error”, and “root mean square error (RMSE)” parameters were used for the performance of the model. The stock crises using the bi-directional LSTM deep neural technique, which has outperformed as 198.21% RMSE value of “Adani-Ports”. The researchers can investigate other technological indications in the future to forecast the crisis point. It can more improved using a new optimizer with a hybrid regression model.