Stock Price Prediction using GRU and Dashboards

Mathematics

The long short-term memory (LSTM) and gated recurrent unit (GRU) models are popular deep-learning architectures for stock market forecasting. Various studies have speculated that incorporating financial news sentiment in forecasting could produce a better performance than using stock features alone. This study carried a normalized comparison on the performances of LSTM and GRU for stock market forecasting under the same conditions and objectively assessed the significance of incorporating the financial news sentiments in stock market forecasting. This comparative study is conducted on the cooperative deep-learning architecture proposed by us. Our experiments show that: (1) both LSTM and GRU are circumstantial in stock forecasting if only the stock market features are used; (2) the performance of LSTM and GRU for stock price forecasting can be significantly improved by incorporating the financial news sentiments with the stock features as the input; (3) both the LSTM-News and GRU-New...

Long term prediction of future prices and returns of financial time series, particularly asset classes such as indices and stocks, has been one of the most challenging problems in both academia and finance. There are many contributing factors to the volatility of stocks and indices prices, such as macroeconomic news, trader psychology, algorithmic trading, and natural disasters such as pandemics, which leads to noisy and non-stationary time series. Although classical statistical prediction models such as ARIMA can predict the next time lag of time series accurately, their performance sharply falls when it comes to predicting longer periods. Besides, they can not generalize beyond a single share or market. As the size, diversity and frequency of financial data evolve, and massive computation power being easily accessible nowadays, we need modern approaches such as machine learning and deep learning to handle the issues of predicting financial time series. LSTM is a variety of in-depth knowledge that has achieved tremendous successes in many fields, such as sounds and speech recognition. LSTMs are also capable of learning from sequence data and addressing problems of RNNs such as vanishing and exploding gradients. This study proposes two LSTM models to predict the close prices and returns of the S&P 500 index and three stocks from the S&P5oo index, including Apple, HP, and Walmart. These models are compared to each other in terms of prediction accuracy with RMSE as the performance measure. Additionally, the prediction accuracies of both LSTM models are compared to the ARIMA(5,1,0) model in predicting close prices. iv Subsequently, this study identifies a proper Cross-validation method particular to financial time series, to address the training data leakage problem Furthermore, this study identifies the most impactful model hyperparameters in the training process. The results of the experiment show the Stacked LSTM model, which incorporates a convolution layer, outperforms the simple LSTM model in predicting both closing prices and returns. Moreover, the result indicates that both LSTM models have achieved a considerably better result in comparison to the ARIMA model. The result of this research affirms that LSTM models can be utilized to predict the future prices of stocks and indices effectively predict future close prices and returns of stocks and indices. The LSTM models trained in this research can be used as software components in various financial applications such as portfolio allocation, risk assessment and algorithmic trading.

International Journal of Intelligent Systems and Applications

Financial time-series prediction has been long and the most challenging issues in financial market analysis. The deep neural networks is one of the excellent data mining approach has received great attention by researchers in several areas of time-series prediction since last 10 years. “Convolutional neural network (CNN) and recurrent neural network (RNN) models have become the mainstream methods for financial predictions. In this paper, we proposed to combine architectures, which exploit the advantages of CNN and RNN simultaneously, for the prediction of trading signals. Our model is essentially presented to financial time series predicting signals through a CNN layer, and directly fed into a gated recurrent unit (GRU) layer to capture long-term signals dependencies. GRU model perform better in sequential learning tasks and solve the vanishing gradients and exploding issue in standard RNNs. We evaluate our model on three datasets for stock indexes of the Hang Seng Indexes (HSI), th...

2020

Prediction of stock groups values has always been attractive and challenging for shareholders. This paper concentrates on the future prediction of stock market groups. Four groups named diversified financials, petroleum, non-metallic minerals and basic metals from Tehran stock exchange are chosen for experimental evaluations. Data are collected for the groups based on ten years of historical records. The values predictions are created for 1, 2, 5, 10, 15, 20 and 30 days in advance. The machine learning algorithms utilized for prediction of future values of stock market groups. We employed Decision Tree, Bagging, Random Forest, Adaptive Boosting (Adaboost), Gradient Boosting and eXtreme Gradient Boosting (XGBoost), and Artificial neural network (ANN), Recurrent Neural Network (RNN) and Long short-term memory (LSTM). Ten technical indicators are selected as the inputs into each of the prediction models. Finally, the result of predictions is presented for each technique based on three metrics. Among all algorithms used in this paper, LSTM shows more accurate results with the highest model fitting ability. Also, for tree-based models, there is often an intense competition between Adaboost, Gradient Boosting and XGBoost.

Alphanumeric Journal, 2023

In this study, LSTM (Long-Short Term Memory) and GRU (Gated Recurrent Unit) techniques of deep learning, which are among the latest advanced technologies, were applied in the Google Colab software program for stock price forecasting. The dataset used in the study was obtained from Yahoo Finance and covers the dates between 02/01/2013 and 30/12/2022. Forecast models were created by considering 5 companies belonging to the XELKT (Electricity Market in Borsa Istanbul) index, which is part of BIST (Borsa Istanbul). Subsequently, the success of these forecast models was tested with the calculated model performance criteria, aiming to determine whether the techniques used were successful in stock price forecasting. Additionally, based on the results of MSE (Mean Squared Error) and MAPE (Mean Absolute Percentage Error) among the calculated model performance criteria, the techniques used were compared with each other, aiming to determine which of these techniques provided forecasts with less error. Then, through the analysis conducted on four different days, an attempt was made to identify the day that yielded the most successful forecasts. As a final step, the goal was to find a model with the least error based on techniques, epoch number, and the number of days forecasted, considering both MSE and MAPE for stocks. Since the model performance criteria outputs obtained from these analyses are below 1 for MSE and below 5% for MAPE, it can be concluded that both techniques demonstrate successful stock price forecasting. Consequently, in the comparison between these two techniques, it is observed that the LSTM technique is slightly more successful than the GRU technique.

Journal of Institute of Science and Technology

The stock price is the cost of purchasing a security or stock in a stock exchange. The stock price prediction has been the aim of investors since the beginning of the stock market. It is the act of forecasting the future price of a company's stock. Nowadays, deep learning techniques are widely used for identifying the stock trends from large amounts of past data. This research has experimented two big and robust commercial banks listed in the Nepal Stock Exchange (NEPSE) and compared stock price prediction performance of GRU with three widely used gradient descent optimization techniques: Momentum, RMSProp, and Adam. GRU with Adam is more accurate and consistent approach for predicting stock prices from the present study.

International Journal of Computing Sciences Research, 2024

Purpose – This paper proposes a novel deep neural network model, specifically long short-term memory (LSTM) networks, for predicting stock prices using historical data and financial indicators. Method – LTSM can handle long sequences while capturing temporal dependencies, making it an excellent choice for NLP or time series. The model is trained and tested on the Ayala Corporation (AYALY) stock dataset from 2016 to 2019, using four financial indicators: earnings per share (EPS), EPS growth, price/earnings ratio, and price/earnings-to-growth ratio. Results – The results show that the model achieves high accuracy and outperforms other Deep Neural Network variants as confirmed by assessing its performance using suitable metrics like mean squared error and mean absolute error. It effectively explored and selected relevant financial indicators, implemented data preprocessing techniques, and trained the model using historical data. Conclusion – The project effectively explored and selected relevant financial indicators and trained LSTM models using historical data, and, thus, met its objectives to develop a deep neural network model for stock price prediction. Recommendations – The authors recommend that future researchers continue to explore the integration of a diverse set of financial indicators, employ rigorous comparative analyses, and experiment with different time frames for future predictions to further enhance prediction accuracy. Research Implications – This paper contributes to the ongoing development of machine-learning studies, especially in the Philippines, particularly for time-series forecasting. With more accurate predictions of stock prices, the study could enable investors to make informed investment decisions, trading strategies, and financial decision-making processes.

Mathematics, 2020

The long short-term memory (LSTM) and gated recurrent unit (GRU) models are popular deep-learning architectures for stock market forecasting. Various studies have speculated that incorporating financial news sentiment in forecasting could produce a better performance than using stock features alone. This study carried a normalized comparison on the performances of LSTM and GRU for stock market forecasting under the same conditions and objectively assessed the significance of incorporating the financial news sentiments in stock market forecasting. This comparative study is conducted on the cooperative deep-learning architecture proposed by us. Our experiments show that: (1) both LSTM and GRU are circumstantial in stock forecasting if only the stock market features are used; (2) the performance of LSTM and GRU for stock price forecasting can be significantly improved by incorporating the financial news sentiments with the stock features as the input; (3) both the LSTM-News and GRU-News models are able to produce better forecasting in stock price equally; (4) the cooperative deep-learning architecture proposed in this study could be modified as an expert system incorporating both the LSTM-News and GRU-News models to recommend the best possible forecasting whichever model can produce dynamically.

Journal of Intelligent Learning Systems and Applications, 2024

A comparative analysis of deep learning models and traditional statistical methods for stock price prediction uses data from the Nigerian stock exchange. Historical data, including daily prices and trading volumes, are employed to implement models such as Long Short Term Memory (LSTM) networks, Gated Recurrent Units (GRUs), Autoregressive Integrated Moving Average (ARIMA), and Autoregressive Moving Average (ARMA). These models are assessed over three-time horizons: short-term (1 year), medium-term (2.5 years), and long-term (5 years), with performance measured by Mean Squared Error (MSE) and Mean Absolute Error (MAE). The stability of the time series is tested using the Augmented Dickey-Fuller (ADF) test. Results reveal that deep learning models, particularly LSTM, outperform traditional methods by capturing complex, nonlinear patterns in the data, resulting in more accurate predictions. However, these models require greater computational resources and offer less interpretability than traditional approaches. The findings highlight the potential of deep learning for improving financial forecasting and investment strategies. Future research could incorporate external factors such as social media sentiment and economic indicators, refine model architectures, and explore real-time applications to enhance prediction accuracy and scalability.

Lecture Notes in Networks and Systems, 2020

The paper proposes a progressive conclusion on the application of recurrent neural networks in stock price forecasting. We have also used random forest classifier to factor in the sudden fluctuations in stock prices which are derivatives of any abnormal events. Machine learning and deep learning strategies are being used by many quantitative hedge funds to increase their returns. Finance data belongs to time series data. A time series is a series of data points indexed in time. The nonlinearity and chaotic nature of the data can be combated using recurrent neural networks which are effective in tracing relationships between historical data and using it to predict new data. Historical data in this context is time series data from the past. It is one of the most important and the most valuable parts for speculating about future prices. Long short-term memory (LSTM) is capable of capturing the most important features from time series data and modelling its dependencies. Building a good and effective prediction system can help investors and traders to get a glimpse of the future direction of the stock and accordingly help them mitigate risk in their respective portfolios. The results obtained by our approach are accurate up to 97% for the values predicted using historic data and 67% for the trend prediction using news headlines.