Downloads

https://doi.org/10.62836/iaet.v1i1.1009
Qian Zhang, & Jiarui Rao. (2022). Enhancing Financial Forecasting Models with Textual Analysis: A Comparative Study of Decomposition Techniques and Sentiment-Driven Predictions. Innovations in Applied Engineering and Technology, 1(1), 1–6. https://doi.org/10.62836/iaet.v1i1.1009
Volume 1, Issue 1

Enhancing Financial Forecasting Models with Textual Analysis: A Comparative Study of Decomposition Techniques and Sentiment-Driven Predictions

Financial time series data are inherently complex, encompassing various components such as trends, seasonal patterns, and irregular fluctuations. This paper presents a comprehensive analysis of decomposition techniques applied to financial time series, aiming to disentangle these components for more accurate forecasting and risk management. We begin by reviewing the traditional methods of decomposition, such as classical decomposition and trend-ratio decomposition, highlighting their limitations in capturing the dynamic nature of financial data. Subsequently, we explore more sophisticated techniques like the Hodrick-Prescott filter, which is widely used for extracting cyclical components, and the Baxter-King filter, which is designed to separate trend and cyclical components while accounting for potential business cycles. The paper then delves into state-of-the-art methods, including the use of autoregressive integrated moving average (ARIMA) models and exponential smoothing state space models, which are capable of capturing both linear and nonlinear patterns in financial time series. We also discuss the application of machine learning algorithms, such as EEMD and long short-term memory (LSTM) networks, which have shown promise in handling the non-stationarity and volatility clustering present in financial data. Empirical analysis is conducted using historical stock prices and exchange rates, demonstrating the effectiveness of these decomposition techniques in isolating the underlying components of financial time series. The results indicate that a combination of traditional and modern methods yields the most robust decomposition, providing a clearer picture of the market dynamics and informing better investment decisions. In conclusion, this paper contributes to the literature by providing a comparative analysis of various decomposition methods and their applicability to financial time series. It underscores the importance of selecting the appropriate technique based on the specific characteristics of the data at hand. The insights gained from this study can be instrumental for financial analysts and economists in developing more effective models for forecasting and risk assessment.

financial time series; decomposition techniques; trend analysis; seasonality; forecasting; risk management

References

  1. Thelwall M, Buckley K, Paltoglou G, Cai D, Kappas A. Sentiment Strength Detection in Short Informal Text. Journal of the American Society for Information Science and Technology 2010; 61(12): 2544–2558.
  2. Melville P, Gryc W, Lawrence RD. Sentiment Analysis of Blogs by Combining Lexical Knowledge with Text Classification. In Proceedings of the 15th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, Paris, France, 28 June–1 July 2009.
  3. Li S, Mo Y, Li Z. Automated Pneumonia Detection in Chest X-ray Images Using Deep Learning Model. Innovations in Applied Engineering and Technology 2022; 1: 1–6.
  4. Wu Z, Wang Q, Andrei V, et al. Pipeline Degradation Evaluation Based on Distributed Fiber Sensors and Convolutional Neural Networks (CNNs). In Proceedings of the 27th International Conference on Optical Fiber Sensors, Technical Digest Series, Optica Publishing Group, Alexandria, VA, USA, 29 August–2 September 2022.
  5. Yadollahi A, Shahraki AG, Zaiane OR. Current State of Text Sentiment Analysis from Opinion to Emotion Mining. ACM Computing Surveys (CSUR) 2017; 50(2): 1–33.
  6. Chen M, Chen Y, Zhang Q. A Review of Energy Consumption in the Acquisition of Bio-Feedstock for Microalgae Biofuel Production. Sustainability 2021; 13(16): 8873.
  7. Sailunaz K, Alhajj R. Emotion and Sentiment Analysis from Twitter Text. Journal of Computational Science 2019; 36: 101003.
  8. Peng Z, Jian J, Wang M, et al. Big Data Analyticson Fiber-Optical Distributed Acoustic Sensing with Rayleigh Enhancements. In Proceedings of the 2019 IEEE Photonics Conference (IPC), San Antonio, TX, USA, 29 September–3 October 2019; pp.1–3.
  9. O'Connor B, Balasubramanyan R, Routledge B, Smith N. From Tweets to Polls: Linking Text Sentiment to Public Opinion Time Series. In Proceedings of the International AAAI Conference on Web and Social Media (Vol. 4, No. 1, pp. 122–129), Washington, DC, USA, 23–26 May 2010.
  10. Wang Q, Zhao K, Badar M, et al. Improving OFDR Distributed Fiber Sensing by Fibers with Enhanced Rayleigh Backscattering and Image Processing. IEEE Sensors Journal 2022; 22(19): 18471–18478. https://doi.org/10.1109/JSEN.2022.3197730.
  11. Badar M, Lu P, Wang M, et al. Integrated Auxiliary Interferometer to Correct Non-Linear Tuning Errors in OFDR. In Proceedings of the SPIE 2020, Online, 27 April–9 May 2020. https://doi.org/10.1117/12.2558910.
  12. Li Y, Shimizu S, Mizumoto M, et al. Proton Beam Therapy for Multifocal Hepatocellular Carcinoma (HCC) Showing Complete Response in Pathological Anatomy After Liver Transplantation. Cureus 2022; 14(6): e25744. https://doi.org/10.7759/cureus.25744.
  13. Kumada H, Li Y, Yasuoka K, et al. Current Development Status of iBNCT001, Demonstrator of a LINAC-based Neutron Source for BNCT. Journal of Neutron Research 2022; 24(3–4): 347–358.
  14. Chen M. Annual Precipitation Forecast of Guangzhou Based on Genetic Algorithm and Backpropagation Neural Network (GA-BP). In Proceedings of the International Conference on Algorithms, High Performance Computing, and Artificial Intelligence (AHPCAI 2021), Sanya, China,19–21 November 2021; Volume 12156, pp. 182–186.
  15. Rout JK, Choo KKR, Dash AK, Bakshi S, Jena SK, Williams KL. A Model for Sentiment and Emotion Analysis of Unstructured Social Media Text. Electronic Commerce Research 2018; 18: 181–199.
  16. Wang Q, Jian J, Wang M, et al. Pipeline Defects Detection and Classification Based on Distributed Fiber Sensors and Neural Networks. In Proceedings of the Optical Fiber Sensors Conference 2020 Special Edition, Washington, DC, USA, 8–12 June 2020.
  17. Dong S, Xu T, Chen M. Solar Radiation Characteristics in Shanghai. Journal of Physics: Conference Series 2020; 2351: 012016.
  18. Ullah MA, Marium SM, Begum SA, Dipa NS. An Algorithm and Method for Sentiment Analysis Using the Text and Emoticon. ICT Express 2020; 6(4): 357–360.
  19. Wang R, Shapiro V. Topological Semantics for Lumped Parameter Systems Modeling. Advanced Engineering Informatics 2019; 42: 100958.
  20. Shimizu S, Mizumoto M, Okumura, et al. Proton Beam Therapy for a Giant Hepatic Hemangioma: A Case Report and Literature Review. Clinical and Translational Radiation Oncology 2021; 27: 152–156. https://doi.org/10.1016/j.ctro.2021.01.014.
  21. Mohammad SM. Sentiment Analysis: Detecting Valence, Emotions, and Other Affectual States from Text. In Emotion Measurement; Woodhead Publishing: Sawston, UK, 2016; pp. 201–237.
  22. Wang R, Behandish M. Surrogate Modeling for Physical Systems with Preserved Properties and Adjustable Tradeoffs. arXiv 2022, arXiv:2202.01139.
  23. Greaves F, Ramirez-Cano D, Millett C, Darzi A, Donaldson L. Use of Sentiment Analysis for Capturing Patient Experience from Free-Text Comments Posted Online. Journal of Medical Internet Research 2013; 15(11): e2721.
  24. Trilla A, Alias F. Sentence-Based Sentiment Analysis for Expressive Text-to-Speech. IEEE Transactions on Audio, Speech, and Language Processing 2012; 21(2): 223–233.

Supporting Agencies

  1. Funding: Not applicable.