KOSPI200 Index Prediction Using Sequence-to-Sequence based on Denoising Filter and Attention Mechanism
0
조회
0
다운로드
4
인용
0
참고문헌
키워드
초록
In the financial market, many studies have been conducted to forecast stock prices, but there are limitations in accurately predicting them due to the high volatility of price and high noise. In addition, in prediction, traditional time series analysis techniques have been widely used, but there are methodological limitations. In this study, the noise of the stock price data is removed by applying a denoising filter, and the prediction performance is then improved using the deep learning long short-term memory (LSTM) model. In addition, as the length of time series data increases, we apply an attention mechanism to predict the KOSPI200 index to minimize the loss of information in the deep learning LSTM model. Four deep learning prediction models were conducted using daily and 30-minute data of the KOSPI200 index according to whether denoising filter and attention mechanisms were applied, and the learning performance was compared and analyzed. As a result, the deep learning LSTM model with both denoising filter and the attention mechanism showed smaller error between the actual and predicted values.
Abstract
In the financial market, many studies have been conducted to forecast stock prices, but there are limitations in accurately predicting them due to the high volatility of price and high noise. In addition, in prediction, traditional time series analysis techniques have been widely used, but there are methodological limitations. In this study, the noise of the stock price data is removed by applying a denoising filter, and the prediction performance is then improved using the deep learning long short-term memory (LSTM) model. In addition, as the length of time series data increases, we apply an attention mechanism to predict the KOSPI200 index to minimize the loss of information in the deep learning LSTM model. Four deep learning prediction models were conducted using daily and 30-minute data of the KOSPI200 index according to whether denoising filter and attention mechanisms were applied, and the learning performance was compared and analyzed. As a result, the deep learning LSTM model with both denoising filter and the attention mechanism showed smaller error between the actual and predicted values.
참고문헌 (15)
- 박종학, "An Investment Strategy Using Genetic Algorithm to Maximize Downside Risk-Adjusted Returns for Long-Term Investors", Quantitative Bio-Science, 36(2), pp.111~117 (2017), 자연과학연구소[DOI]
- 고한준, "Segmentation Strategy for Asset Groups in the Korean Exchange-Traded Funds Market Using Genetic Algorithm", Quantitative Bio-Science, 38(2), pp.87~92 (2019), 자연과학연구소[DOI]
- 황선영, "Volatility for High Frequency Time Series Toward fGARCH (1,1) as a Functional Model", Quantitative Bio-Science, 37(2), pp.73~79 (2018), 자연과학연구소[DOI]
- Long W, "Deep learning-based feature engineering for stock price movement prediction", Knowl-Based Syst, 164, pp.163~173 (2018)
- 김성수, "순환 신경망 기술을 이용한 코스피 200 지수에 대한 예측 모델 개발 및 성능 분석 연구", 한국산업정보학회논문지, 22(6), pp.23~29 (2017), 한국산업정보학회[DOI]
- 하대우, "XGBoost 모형을 활용한 코스피 200 주가지수 등락 예측에 관한 연구", 한국데이터정보과학회지, 30(3), pp.655~669 (2019), 한국데이터정보과학회[DOI]
- 이낙영, "디노이징 필터와 LSTM을 활용한 KOSPI200 선물지수 예측", 한국데이터정보과학회지, 30(3), pp.645~654 (2019), 한국데이터정보과학회[DOI]
- 이동원, "KOSPI200 Prediction through Low-Pass Filtered Long Short-Term Memory Algorithm", Quantitative Bio-Science, 39(1), pp.25~31 (2020), 자연과학연구소[DOI]
- LIU Y, "Applications of Savitzky-Golay filter for seismic random noise reduction", Inst Geopyysics Pas, 64, pp.101~124 (2016)
- Chen J, "A simple method for reconstructing a high-quality NDVI timeseries data set based on the Savitzky-Golay filter", Remote Sens Environ, 91, pp.332~344 (2004)
- Li M, "Reconstructing vegetation temperature condition index based on the Savitzky-Golay filter", IFIP AICT, 346, pp.629~637 (2011)
- Hochreiter S, "Long short-term memory", Neural Comput, 9, pp.1735~1780 (1997)
- Cho K, "Learning phrase representations using RNN Encoder-Decoder for statistical machine translation", arXiv preprint 2014;arXiv:1406.1078, ~
- Bahdanau D, "Neural machine translation by jointly learning to align and translate", arXiv preprint 2014;arXiv:1409. 0473, ~
- 김은총, "은닉 마르코프 모델 (HMM)과 유전자 알고리즘 (GA)를 이용한 자산배분전략 제안", 한국데이터정보과학회지, 30(1), pp.33~44 (2019), 한국데이터정보과학회[DOI]