Genetic Algorithm-Optimized Long Short-Term Memory Network for Stock Market Prediction

Chung, Hye-Jung; Shin, Kyung Shik

doi:10.3390/su10103765

Cited by 234 publications

(167 citation statements)

References 54 publications

Supporting

Mentioning

158

Contrasting

Unclassified

Order By: Relevance

“…In recent years, with the significant advances in deep learning techniques, recurrent neural networks have emerged as a promising model for handling sequential data in various tasks such as natural language processing, speech recognition, and computer vision. Moreover, several studies proved that RNN including LSTM cells is the most useful model in the financial time series prediction problem [30][31][32][33][34][35][36]. For example, Chen et al [30] introduced a stock price movement prediction model based on LSTM.…”

Section: Stock Market Predictionmentioning

confidence: 99%

See 1 more Smart Citation

A Novel Approach to Short-Term Stock Price Movement Prediction using Transfer Learning

Nguyen

Yoon

2019

Applied Sciences

View full text Add to dashboard Cite

Stock price prediction has always been an important application in time series predictions. Recently, deep neural networks have been employed extensively for financial time series tasks. The network typically requires a large amount of training samples to achieve high accuracy. However, in the stock market, the number of data points collected on a daily basis is limited in one year, which leads to insufficient training samples and accordingly results in an overfitting problem. Moreover, predicting stock price movement is affected by various factors in the stock market. Therefore, choosing appropriate input features for prediction models should be taken into account. To address these problems, this paper proposes a novel framework, named deep transfer with related stock information (DTRSI), which takes advantage of a deep neural network and transfer learning. First, a base model using long short-term memory (LSTM) cells is pre-trained based on a large amount of data, which are obtained from a number of different stocks, to optimize initial training parameters. Second, the base model is fine-tuned by using a small amount data from a target stock and different types of input features (constructed based on the relationship between stocks) in order to enhance performance. Experiments are conducted with data from top-five companies in the Korean market and the United States (US) market from 2012 to 2018 in terms of the highest market capitalization. Experimental results demonstrate the effectiveness of transfer learning and using stock relationship information in helping to improve model performance, and the proposed approach shows remarkable performance (compared to other baselines) in terms of prediction accuracy.

show abstract

Section: Stock Market Predictionmentioning

confidence: 99%

“…They concluded that LSTM could attain satisfactory performance in predicting China's index price movements. Chung and Shin [34] employed an LSTM network to predict KOSPI index prices in the next day. They tested their method on KOSPI index data from 2000 to 2016 and found that the method achieved satisfactory performance.…”

Section: Stock Market Predictionmentioning

confidence: 99%

A Novel Approach to Short-Term Stock Price Movement Prediction using Transfer Learning

Nguyen

Yoon

2019

Applied Sciences

View full text Add to dashboard Cite

show abstract

“…Unlike LTMA, LSTM introduces cells and gates to form a "highway" to retain gradient information in a long sequence of a recurrent neural network. In [46], a genetic algorithm was integrated with an LSTM network to optimize time window size and architectural factors, to better predict the Korea Composite Stock Price Index.…”

Section: Related Workmentioning

confidence: 99%

Long Term Memory Assistance for Evolutionary Algorithms

et al. 2019

View full text Add to dashboard Cite

Short term memory that records the current population has been an inherent component of Evolutionary Algorithms (EAs). As hardware technologies advance currently, inexpensive memory with massive capacities could become a performance boost to EAs. This paper introduces a Long Term Memory Assistance (LTMA) that records the entire search history of an evolutionary process. With LTMA, individuals already visited (i.e., duplicate solutions) do not need to be re-evaluated, and thus, resources originally designated to fitness evaluations could be reallocated to continue search space exploration or exploitation. Three sets of experiments were conducted to prove the superiority of LTMA. In the first experiment, it was shown that LTMA recorded at least 50 % more duplicate individuals than a short term memory. In the second experiment, ABC and jDElscop were applied to the CEC-2015 benchmark functions. By avoiding fitness re-evaluation, LTMA improved execution time of the most time consuming problems F 03 and F 05 between 7% and 28% and 7% and 16%, respectively. In the third experiment, a hard real-world problem for determining soil models’ parameters, LTMA improved execution time between 26% and 69%. Finally, LTMA was implemented under a generalized and extendable open source system, called EARS. Any EA researcher could apply LTMA to a variety of optimization problems and evolutionary algorithms, either existing or new ones, in a uniform way.

show abstract

“…Moreover, to add realistic assumptions to the current stock, such as transaction costs, liquidity issues, and bid and spread basis, Meng and Khushi [38] suggested using a reinforcement learning model in financial markets to enhance the current models in the field. Moreover, optimization algorithms are used with artificial intelligent prediction models to improve the efficiency of the predications in addition to enhancing the computation complexity [39,40]. Moreover, many methods in the literature suggested using different models to improve the prediction model of financial markets, namely neural network [41] and Markov chain [42] in addition to evaluating different machine learning models as discussed in [43].…”

Section: Introductionmentioning

confidence: 99%

Can International Market Indices Estimate TASI’s Movements? The ARIMA Model

Assous

Al-Rousan

AL-Najjar

et al. 2020

Journal of Open Innovation: Technology, Market, and Complexity

View full text Add to dashboard Cite

This study investigates the effectiveness of six of the key international indices in estimating Saudi financial market (TADAWUL) index (TASI) movement. To investigate the relationship between TASI and other variables, six equations were built using two independent variables of time and international index, while TASI was the dependent variable. Linear, logarithmic, quadratic, cubic, power, and exponential equations were separately used to achieve the targeted results. The results reveal that power equation is the best equation for forecasting the TASI index with a low error rate and high determination coefficient. Additionally, findings of the AutoRegressive Integrated Moving Average (ARIMA) model represent the most important variables to use in order to build a prediction model that can estimate the TASI index. The ARIMA model (with Expert Modeler) coefficients are described as ARIMA (0,1,14). The results show that the SP500, NIKKEI, CAC40, and HSI indices are the most suitable variables for estimating TASI with an R2 and RMSE equal to 0.993 and 113, respectively. This relationship can be used on the previous day to estimate the opening price of TASI based on the closing prices of international indices.

show abstract

Genetic Algorithm-Optimized Long Short-Term Memory Network for Stock Market Prediction

Cited by 234 publications

References 54 publications

A Novel Approach to Short-Term Stock Price Movement Prediction using Transfer Learning

A Novel Approach to Short-Term Stock Price Movement Prediction using Transfer Learning

Long Term Memory Assistance for Evolutionary Algorithms

Can International Market Indices Estimate TASI’s Movements? The ARIMA Model

Contact Info

Product

Resources

About