Accurate prediction of future prices of stocks is a difficult task to perform. Even more challenging is to design an optimized portfolio with weights allocated to the stocks in a way that optimizes its return and the risk. This paper presents a systematic approach towards building two types of portfolios, optimum risk, and eigen, for four critical economic sectors of India. The prices of the stocks are extracted from the web from Jan 1, 2016, to Dec 31, 2020. Sector-wise portfolios are built based on their ten most significant stocks. An LSTM model is also designed for predicting future stock prices. Six months after the construction of the portfolios, i.e., on Jul 1, 2021, the actual returns and the LSTM-predicted returns for the portfolios are computed. A comparison of the predicted and the actual returns indicate a high accuracy level of the LSTM model.

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.

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.

No abstract

No abstract