2021
DOI: 10.48550/arxiv.2107.09055
|View full text |Cite
Preprint
|
Sign up to set email alerts
|

Stock price prediction using BERT and GAN

Abstract: The stock market has been a popular topic of interest in the recent past. The growth in the inflation rate has compelled people to invest in the stock and commodity markets and other areas rather than saving. Further, the ability of Deep Learning models to make predictions on the time series data has been proven time and again. Technical analysis on the stock market with the help of technical indicators has been the most common practice among traders and investors. One more aspect is the sentiment analysis -th… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

0
15
0

Year Published

2022
2022
2024
2024

Publication Types

Select...
5
2
1

Relationship

0
8

Authors

Journals

citations
Cited by 10 publications
(15 citation statements)
references
References 25 publications
0
15
0
Order By: Relevance
“…Combining the sentiment information extracted from BERT on carefully annotated Twitter data with stock OHLC price data, Dong, Yan, Almudaifer, Yan, Jiang and Zhou (2020) reported better results in terms of AUC for next-day price prediction compared to the state-of-the-art stock prediction model StockNet developed by (Xu and Cohen, 2018). Other similar work using BERT equally reported that their models outperformed others to various degrees (Sonkiya, Bajpai and Bansal, 2021;Chen, 2021). Based on these results, we have opted to use the state-of-the-art BERT model for this research, pretrained on a financial lexicon.…”
Section: Stock Price Prediction With Nlpmentioning
confidence: 98%
“…Combining the sentiment information extracted from BERT on carefully annotated Twitter data with stock OHLC price data, Dong, Yan, Almudaifer, Yan, Jiang and Zhou (2020) reported better results in terms of AUC for next-day price prediction compared to the state-of-the-art stock prediction model StockNet developed by (Xu and Cohen, 2018). Other similar work using BERT equally reported that their models outperformed others to various degrees (Sonkiya, Bajpai and Bansal, 2021;Chen, 2021). Based on these results, we have opted to use the state-of-the-art BERT model for this research, pretrained on a financial lexicon.…”
Section: Stock Price Prediction With Nlpmentioning
confidence: 98%
“…They showed that using the textual information significantly boosted the performance of intraday stock trading models. There were also studies using other approaches to incorporating sentiment features, such as Sonkiya et al [30], which leverages news sentiments from BERT and feeds these textual features to Generative Adversarial Networks [12], as well as Chen et el. [7], which makes use of contextualized embeddings generated from news headlines for price prediction.…”
Section: Related Workmentioning
confidence: 99%
“…Faraz and Khaloozadeh [6] enhance GAN with wavelet transformation and z-score method in the data preprocessing phase. Sonkiya et al [45] generate binary sentiment score by FinBERT and employ GAN for stock price predictions. Such methods adapt Vanilla GAN which often suffers from mode collapse, and neglect possible driving factors such as complex public emotions, financial policies, and events, which limits the ability to achieve superior performance.…”
Section: Generative Adversarial Nets On Time Seriesmentioning
confidence: 99%