Stock price prediction with optimized deep LSTM network with artificial rabbits optimization algorithm

The stock market is a financial market where shares of publicly listed corporations are purchased and sold. It is an indicator of a country's economic health, reflecting the performance of companies and the overall business environment. The prices of stocks are determined by supply and demand. Investing in the stock market can be risky, but it can offer the potential for significant returns over the long term. Artificial intelligence, including the stock market, has become increasingly prevalent in the financial sector. Long Short-Term Memory (LSTM) is a type of artificial neural network that is often used in time series analysis. It can effectively predict stock market prices by handling data with multiple input and output timesteps. Metaheuristic algorithms, such as Artificial Rabbits Optimization algorithm (ARO), can be used to optimize the hyperparameters of an LSTM model and improve the accuracy of stock market predictions. In this paper, an optimized deep LSTM network with the ARO model (LSTM-ARO) is created to predict stock prices. DJIA index stocks are used as the dataset. LSTM-ARO is compared with one artificial neural network (ANN) model, three different LSTM models, and LSTM optimized by Genetic Algorithm (GA) model. All the models are tested on MSE, MAE, MAPE, and R2 evaluation criteria. The results show that LSTM-ARO overcomes the other models.