Stock Price Prediction Using Deep Learning Models

Abstract

Prediction of future movement of stock prices has been a subject matter of many research work. On one hand, we have proponents of the Efficient Market Hypothesis who claim that stock prices cannot be predicted, on the other hand, there are propositions illustrating that, if appropriately modeled, stock prices can be predicted with a high level of accuracy. There is also a gamut of literature on technical analysis of stock prices where the objective is to identify patterns in stock price movements and profit from it. In this work, we propose a hybrid approach for stock price prediction using five deep learning-based regression models. We select the NIFTY 50 index values of the National Stock Exchange (NSE) of India, over a period of December 29, 2014 to July 31, 2020. Based on the NIFTY data during December 29, 2014 to December 28, 2018, we build two regression models using <i>convolutional neural networks</i> (CNNs), and three regression models using <i>long-and-short-term memory</i> (LSTM) networks for predicting the <i>open</i> values of the NIFTY 50 index records for the period December 31, 2018 to July 31, 2020. We adopted a multi-step prediction technique with <i>walk-forward validation</i>. The parameters of the five deep learning models are optimized using the grid-search technique so that the validation losses of the models stabilize with an increasing number of epochs in the model training, and the training and validation accuracies converge. Extensive results are presented on various metrics for all the proposed regression models. The results indicate that while both CNN and LSTM-based regression models are very accurate in forecasting the NIFTY 50 <i>open</i> values, the CNN model that previous one week’s data as the input is the fastest in its execution. On the other hand, the encoder-decoder convolutional LSTM model uses the previous two weeks’ data as the input is found to be the most accurate in its forecasting results.

Text2Price: Adeep Learning Model for Predicting Electronic Product Prices from Descriptive Text Sequences

“…Moreover, the effectiveness of a predictive model relies on the selection of variables incorporated in constructing the model, the algorithms employed, and the optimization techniques applied The researchers introduced ve predictive models for forecasting stock prices-two models based on convolutional neural networks (CNNs) and three models based on long short-term memory (LSTM) networks. [7] Future research could explore multistepcommodity price forecasting by adjusting the incorporated data to create a more extensive dataset.…”

Machine learning techniques and methodology analysis for stock market price prediction

“…The NB is a machine learning algorithm used to classify meshy datasets that creates the Bayesian networks based on probability for a specific Bayes-based dataset theorem [39,40] . It works on independent features and does not work with dependent variables.…”

Research on Ensemble Learning-Based Feature Selection Method for Time-Series Prediction

“…To overcome the limitations of linear methods, subsequent research proposed the integration of nonlinear models to enhance the comprehension of complex data, thereby giving rise to nonlinear prediction methods. Prominent among these methods are BP neural networks [6][7][8], support vector machines, recurrent neural networks [9][10][11], generative adversarial networks [12,13], and reinforcement learning [14][15][16]. By employing these methodologies, researchers have achieved a more comprehensive capture of the nonlinear relationships embedded in financial time-series data, leading to relatively accurate prediction outcomes.…”