StockGAN: robust stock price prediction using GAN algorithm

Diqi, Mohammad; Hiswati, Marselina Endah; Nur, Adri Saputra

doi:10.1007/s41870-022-00929-6

Cited by 19 publications

(5 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The performance of this model was better than that of the standard model (Lin et al, 2021). Diqi et al (2022) proposed a GAN model and gave 𝑅 2 real = 0.811166, and 𝑅 2 synthetic = 0.674971. Shahi et al (2020) compared LSTM and GRU to include financial news sentiment.…”

Section: Introductionmentioning

confidence: 93%

“…Among the results of several previous studies, EMAGRU can be a reliable alternative for predicting stock prices. (Ji et al, 2021) 0.1100 0.0190 0.9570 CNN-LSTM (Lu et al, 2020) 0.3969 0.2756 0.9646 WLSTM+Att (Qiu et al, 2020) 0.1971 0.1569 0.9621 LSTM-BO-XGB (Liwei et al, 2021) 0.0610 0.1560 LSTM (Nabipour et al, 2020) 0.0065 0.4300 0.4460 LSTM BGA8X13 (Wang et al, 2021) 1.1345 0.0020 LSTM (DVA) (Ta et al, 2020) 0.0597 LASSO-LSTM (Gao et al, 2021) 0.2709 0.1880 PCA-LSTM (Gao et al, 2021) 0.6368 0.4289 GAN (Diqi et al, 2022) 0.0207 0.8112 GAN (Lin et al, 2021) 0.0533 GRU-Only (Shahi et al, 2020) 0.4731 0.4281 0.8790 GRU-News (Shahi et al, 2020) 0.2915 0.2447 0.9670 LASSO-GRU (Gao et al, 2021) 0.2815 0.1986 PCA-GRU (Gao et al, 2021) 0.8121 0.6132 GRU (DVA) (Ta et al, 2020) 0.0473 GRU (S&P500) (Zhang & Fang, 2021) 0.0843 0.0710 GRU (Manjunath et al, 2021) 0.1200 0.1060 Proposed EMAGRU 0.0050 0.0058 0.0045 0.9982…”

Section: Comparison To the Previous Studiesmentioning

confidence: 99%

See 1 more Smart Citation

Performance Evaluation of Stock Prediction Models Using Emagru

ERIZAL,

DIQI

2023

Appl. Comput. Sci.

Self Cite

View full text Add to dashboard Cite

Stock prediction is an exciting issue and is very much needed by investors and business people to develop their assets. The main difficulties in predicting stock prices are dynamic movements, high volatility, and noises caused by company performance and external influences. The traditional method used by investors is the technical analysis based on statistics, valuation of previous stock portfolios, and news from the mass media and social media. Deep learning can predict stock price movements more accurately than traditional methods. As a solution to the issue of stock prediction, we offer the Exponential Moving Average Gated Recurrent Unit (EMAGRU) model and demonstrate its utility. The EMAGRU architecture contains two stacked GRUs arranged in parallel. The inputs and outputs are the EMA10 and EMA20, formed from the closing prices over ten years. We also combine the AntiReLU and ReLU activation functions into the model so that EMAGRU has 6 model variants. Our proposed model produced low losses and high accuracy. RMSE, MEPA, MAE, R2 and were 0.0060, 0.0064, 0.0050, and 0.9976 for EMA10, and 0.0050, 0.0058, 0.0045, and 0.9982 for EMA20, respectively.

show abstract

Section: Introductionmentioning

confidence: 93%

Section: Comparison To the Previous Studiesmentioning

confidence: 99%

Performance Evaluation of Stock Prediction Models Using Emagru

ERIZAL,

DIQI

2023

Appl. Comput. Sci.

Self Cite

View full text Add to dashboard Cite

show abstract

“…The experimental results demonstrated that GAN-based portfolio returns and Sharpe ratios significantly surpassed those of the Fama-French model in the U.S. market. Diqi et al [ 52 ] introduced Stock-GAN, leveraging the GAN algorithm for robust stock price prediction. Their model utilized features such as date, open, high, low, close, and volume of stocks in Indonesia market.…”

Section: Related Literaturementioning

confidence: 99%

Factor-GAN: Enhancing stock price prediction and factor investment with Generative Adversarial Networks

Wang,

Chen

2024

PLoS ONE

View full text Add to dashboard Cite

Deep learning, a pivotal branch of artificial intelligence, has increasingly influenced the financial domain with its advanced data processing capabilities. This paper introduces Factor-GAN, an innovative framework that utilizes Generative Adversarial Networks (GAN) technology for factor investing. Leveraging a comprehensive factor database comprising 70 firm characteristics, Factor-GAN integrates deep learning techniques with the multi-factor pricing model, thereby elevating the precision and stability of investment strategies. To explain the economic mechanisms underlying deep learning, we conduct a subsample analysis of the Chinese stock market. The findings reveal that the deep learning-based pricing model significantly enhances return prediction accuracy and factor investment performance in comparison to linear models. Particularly noteworthy is the superior performance of the long-short portfolio under Factor-GAN, demonstrating an annualized return of 23.52% with a Sharpe ratio of 1.29. During the transition from state-owned enterprises (SOEs) to non-SOEs, our study discerns shifts in factor importance, with liquidity and volatility gaining significance while fundamental indicators diminish. Additionally, A-share listed companies display a heightened emphasis on momentum and growth indicators relative to their dual-listed counterparts. This research holds profound implications for the expansion of explainable artificial intelligence research and the exploration of financial technology applications.

show abstract

“…So, the seller can benefit from the compensation provided by the seller. This compensation can also be generated from exchanging goods and services between the parties [6] [18].…”

Section: Related Workmentioning

confidence: 99%

Effective Stock Prediction Model Using MACD Method

Hamzah

Winardi²

2022

ijicom

View full text Add to dashboard Cite

Stock market predictions help investors to optimize benefits in the financial markets. Various papers have proposed different techniques in stock market forecasting, but no model can provide accurate predictions. In this study, we discuss how to predict stock prices using a MACD (Moving Average Convergence/Divergence Oscillator) method. We collect the dataset, preprocess it, extract features, evaluate the model, and then deploy the MACD method to develop a stock price prediction model. In this study, we collect several features, including date, open, high, low, close, and volume, to conduct the training and testing process. The results of the experiments reveal good accuracy and a low error rate. As a result, it has the potential to be a promising solution for dealing with accurate and dynamic prices. Based on the experimental result, our proposed model can obtain a transaction profit rate of 40.00% and an average profit per transaction of 1.42%.

show abstract

StockGAN: robust stock price prediction using GAN algorithm

Cited by 19 publications

References 24 publications

Performance Evaluation of Stock Prediction Models Using Emagru

Performance Evaluation of Stock Prediction Models Using Emagru

Factor-GAN: Enhancing stock price prediction and factor investment with Generative Adversarial Networks

Effective Stock Prediction Model Using MACD Method

Contact Info

Product

Resources

About