BERT를 활용한 뉴스 기사 감성분석과 블랙-리터만 모형을 결합한 자산 배분 전략 제안
Asset Allocation Strategy based on News Article Sentiment Analysis using BERT and Black-Litterman Model
0
조회
0
다운로드
2
인용
0
참고문헌
키워드
초록
본 연구에서는 투자자의 주관적인 판단을 최소화하기 위해 뉴스 기사에 내포된 감성정보를 딥러닝 모형을 활용하여 추출한 후, 이를 블랙-리터만 모형에 적용하는 객관적인 자산 배분 전략을 제시한다. 연구 결과, 뉴스 기사로부터 추출한 감성을 자산별 기대수익에 대한 전망에 활용하여 포트폴리오를 구성하는 경우 시장 포트폴리오, 동일비중 포트폴리오, 평균분산 포트폴리오보다 수익성이 개선된 것을 확인하였다. 특히 감성이 긍정도 부정도 아닌 중립인 기사를 제외할 경우 수익성이 더욱 개선된다는 점을 보였으며, 통합된 감성정보의 극성을 반영하여 포트폴리오를 구성할 경우 강도를 반영하는 경우보다 자산 배분의 수익성이 개선되는 것을 확인하였다. 본 연구는 자산 배분 전략을 구성할 때 기대수익에 대한 투자자의 주관적인 판단을 활용함에 따라 발생할 수 있는 한계를 극복하는 방안으로서 대중에게 공개된 뉴스 기사의 객관적인 정보를 적용하는 자산 배분 전략 구성 방법을 제안하고, 그 활용 가능성과 우수성을 확인하였다는 측면에서 의의를 가진다.
Abstract
This study introduces an asset allocation strategy that utilize sentiment analysis from news articles through a deep learning model. The derived sentiments are then integrated into the Black-Litterman model, offering a systematic approach to mitigate the inherent subjectivity in investor decisions. Empirical findings from this study reveal the superiority of this proposed approach compared with conventional benchmark portfolios such as market, equal-weighted, and mean-variance portfolios. In particular, the exclusion of articles generating neutral sentiment (neither positive nor negative) further enhances the profitability of the portfolio. In addition, the study shows that constructing portfolios based on the polarity of sentiment, rather than considering its positive or negative intensity , improves profitability. The significance of this study lies in its introduction of a novel framework for constructing asset allocation strategies. By utilizing objective information from publicly available news articles, it effectively circumvents the limitations tied to subjective investor judgment in predicting expected returns. The demonstrated feasibility and superiority of this data-driven approach in asset allocation strategies underscore its potential to revolutionize current practices.
참고문헌 (31)
- Alexander, D., "Application of Ensemble Learning for Views Generation in Meucci Portfolio Optimization Framework", Review of Business and Economics Studies, 1, pp.100~110 (2013)
- Araci, D., "Finbert: Financial Sentiment Analysis with pre-trained Language Models", arXiv preprint arXiv:1908.10063, ~ (2019)
- Baker, H. K., "How Biases Affect Investor Behaviour", European Financial Review, pp.7~10 (2014)
- Beach, S. L., "An Application of the Black-Litterman Model with EGARCH-M-derived Views for International Portfolio Management", Financial Markets and Portfolio Management, 21, pp.147~166 (2007)
- Best, M. J., "On the Sensitivity of Mean-variance-efficient Portfolios to Changes in Asset Means : Some Analytical and Computational Results", Review of Financial Studies, 4(2), pp.315~342 (1991)
- Black, F., "Asset Allocation : Combining Investor Views with Market Equilibrium", Goldman Sachs Fixed Income Research, 115(1), pp.7~18 (1990)
- Britten-Jones, M., "The Sampling Error in Estimates of Mean-variance Efficient Portfolio Weights", Journal of Finance, 54(2), pp.655~671 (1999)
- Daniel, K., "Investor Psychology and Security Market Under-and Overreactions", Journal of Finance, 53(6), pp.1839~1885 (1998)
- Daniel, K., "Investor Psychology in Capital Markets : Evidence and Policy Implications", Journal of Monetary Economics, 49(1), pp.139~209 (2002)
- Devlin, J., "Bert : Pre-training of Deep Bidirectional Transformers for Language Understanding", arXiv preprint arXiv : 1810. 04805, ~ (2018)
- Drobetz, W., "How to Avoid the Pitfalls in Portfolio Optimization? Putting the Black-Litterman Approach at Work", Financial Markets and Portfolio Management, 15(1), pp.59~ (2001)
- He, G., "The Intuition Behind Black-Litterman Model Portfolios", ~ (2002)
- Idzorek, T., "Forecasting Expected Returns in The Financial Markets", pp.17~38 (2005), Academic Press
- Jobson, J. D., "Putting Markowitz Theory to Work", Journal of Portfolio Management, 7(4), pp.70~74 (1981)
- Kara, M., "A Hybrid Approach for Generating Investor Views in Black–Litterman Model", Expert Systems with Applications, 128, pp.256~270 (2019)
- Kolm, P. N., "Black-Litterman and Beyond : The Bayesian Paradigm in Investment Management", Journal of Portfolio Management, 47(5), pp.91~113 (2021)
- Li, X., "News Impact on Stock Price Return Via Sentiment Analysis", Knowledge-Based Systems, 69, pp.14~23 (2014)
- Mahajan, A., "Mining Financial News for Major Events and Their Impacts on the Market", IEEE, 1~ (2008)
- Malo, P., "Good Debt or Bad Debt : Detecting Semantic Orientations in Economic Texts", Journal of the Association for Information Science and Technology, 65(4), pp.782~796 (2014)
- Markowitz, H., "Portfolio Selection", Journal of Finance, 7(1), pp.77~91 (1952)
- Meucci, A., "The Black-litterman Approach: Original Model and Extensions", Encyclopedia of Quantitative Finance (2010)
- Michaud, R. O., "The Markowitz Optimization Enigma : Is ‘Optimized’ Optimal?", Financial Analysts Journal, 45(1), pp.31~42 (1989)
- Peters, M. E., "Semi-supervised Sequence Tagging with Bidirectional Language Models", arXiv preprint arXiv : 1705. 00108, ~ (2017)
- Radford, A., "Improving Language Understanding by Generative Pre-training", ~ (2018)
- Shah, D., "Stock Market Analysis : A Review and Taxonomy of Prediction Techniques", International Journal of Financial Studies, 7(2), pp.26~ (2019)
- Sokolov, A., "Weak Supervision and Black-litterman for Automated Esg Portfolio Construction", Journal of Financial Data Science, 3(3), pp.129~138 (2021)
- Sonkiya, P., "Stock Price Prediction using BERT and GAN", arXiv preprint arXiv : 2107. 09055, ~ (2021)
- Sousa, M. G., "BERT for Stock Market Sentiment Analysis", IEEE, ~ (2019)
- Vaswani, A., "Attention is All You Need", Advances in Neural Information Processing Systems, 30 (2017)
- Walters, J., "The Black-Litterman Model in Detail", ~ (2014)
- Xing, F. Z., "Intelligent Asset Allocation Via Market Sentiment Views", 13(4), pp.25~34 (2018), IEEE