Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
169 tokens/sec
GPT-4o
7 tokens/sec
Gemini 2.5 Pro Pro
45 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
38 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Enhanced LFTSformer: A Novel Long-Term Financial Time Series Prediction Model Using Advanced Feature Engineering and the DS Encoder Informer Architecture (2310.01884v2)

Published 3 Oct 2023 in cs.LG and cs.AI

Abstract: This study presents a groundbreaking model for forecasting long-term financial time series, termed the Enhanced LFTSformer. The model distinguishes itself through several significant innovations: (1) VMD-MIC+FE Feature Engineering: The incorporation of sophisticated feature engineering techniques, specifically through the integration of Variational Mode Decomposition (VMD), Maximal Information Coefficient (MIC), and feature engineering (FE) methods, enables comprehensive perception and extraction of deep-level features from complex and variable financial datasets. (2) DS Encoder Informer: The architecture of the original Informer has been modified by adopting a Stacked Informer structure in the encoder, and an innovative introduction of a multi-head decentralized sparse attention mechanism, referred to as the Distributed Informer. This modification has led to a reduction in the number of attention blocks, thereby enhancing both the training accuracy and speed. (3) GC Enhanced Adam & Dynamic Loss Function: The deployment of a Gradient Clipping-enhanced Adam optimization algorithm and a dynamic loss function represents a pioneering approach within the domain of financial time series prediction. This novel methodology optimizes model performance and adapts more dynamically to evolving data patterns. Systematic experimentation on a range of benchmark stock market datasets demonstrates that the Enhanced LFTSformer outperforms traditional machine learning models and other Informer-based architectures in terms of prediction accuracy, adaptability, and generality. Furthermore, the paper identifies potential avenues for future enhancements, with a particular focus on the identification and quantification of pivotal impacting events and news. This is aimed at further refining the predictive efficacy of the model.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (54)
  1. \bibinfoauthorRuey S. Tsay, \bibinfoyear2005. \bibinfotitleAnalysis of financial time series. \bibinfopublisherJohn Wiley & Sons.
  2. \bibinfotitleThe (mis) behaviour of markets: a fractal view of risk, ruin and reward. \bibinfopublisherProfile books.
  3. \bibinfotitleIntroduction to time series and forecasting. \bibinfopublisherSpringer.
  4. \bibinfotitleThe empirical mode decomposition and the Hilbert spectrum for nonlinear and non-stationary time series analysis. \bibinfojournalProceedings of the Royal Society of London. Series A: mathematical, physical and engineering sciences \bibinfovolume454(1971), \bibinfopages903–995. \bibinfopublisherThe Royal Society.
  5. \bibinfotitleWind Power Generation Forecast Based on Multi-Step Informer Network. \bibinfojournalEnergies \bibinfovolume15(18), \bibinfopages6642. \bibinfopublisherMDPI.
  6. \bibinfotitleImproving the accuracy of multi-step prediction of building energy consumption based on EEMD-PSO-Informer and long-time series. \bibinfojournalComputers and Electrical Engineering \bibinfovolume110, \bibinfopages108845. \bibinfopublisherElsevier.
  7. \bibinfotitleLoad forecasting of district heating system based on Informer. \bibinfojournalEnergy \bibinfovolume253, \bibinfopages124179. \bibinfopublisherElsevier.
  8. \bibinfotitleFinancial Analysis, Planning, and Forecasting. In: \bibinfobooktitleEssentials of Excel VBA, Python, and R: Volume II: Financial Derivatives, Risk Management and Machine Learning, \bibinfopages433–455. \bibinfopublisherSpringer.
  9. \bibinfotitleEffects of the validation set on stock returns forecasting. \bibinfojournalExpert Systems with Applications \bibinfovolume150, \bibinfopages113271. \bibinfopublisherElsevier.
  10. \bibinfotitleStock market prediction via multi-source multiple instance learning. \bibinfojournalIEEE Access \bibinfovolume6, \bibinfopages50720–50728. \bibinfopublisherIEEE.
  11. \bibinfotitleA comprehensive comparative study of artificial neural network (ANN) and support vector machines (SVM) on stock forecasting. \bibinfojournalAnnals of Data Science \bibinfovolume10(\bibinfonumber1), \bibinfopages183–208. \bibinfopublisherSpringer.
  12. Omer Berat Sezer, Ahmet Murat Ozbayoglu, ”Algorithmic financial trading with deep convolutional neural networks: Time series to image conversion approach”, Applied Soft Computing, 70, 525–538, 2018. Elsevier.
  13. \bibinfotitleChatGPT: Unlocking the future of NLP in finance. \bibinfojournalAvailable at SSRN 4323643.
  14. \bibinfoauthorRobert F. Engle, \bibinfoyear1982. \bibinfotitleAutoregressive conditional heteroscedasticity with estimates of the variance of United Kingdom inflation. \bibinfojournalEconometrica: Journal of the econometric society \bibinfovolume, \bibinfopages987–1007. \bibinfopublisherJSTOR.
  15. \bibinfoauthorTim Bollerslev, \bibinfoyear1986. \bibinfotitleGeneralized autoregressive conditional heteroskedasticity. \bibinfojournalJournal of econometrics \bibinfovolume31(\bibinfonumber3), \bibinfopages307–327. \bibinfopublisherElsevier.
  16. \bibinfoauthorDavid A. Hsieh, \bibinfoyear1988. \bibinfotitleThe statistical properties of daily foreign exchange rates: 1974–1983. \bibinfojournalJournal of international economics \bibinfovolume24(1-2), \bibinfopages129–145. \bibinfopublisherElsevier.
  17. \bibinfoauthorRobert Goodell Brown, \bibinfoyear2004. \bibinfotitleSmoothing, forecasting and prediction of discrete time series. \bibinfopublisherCourier Corporation.
  18. \bibinfotitleAsian stock markets closing index forecast based on secondary decomposition, multi-factor analysis and attention-based LSTM model. \bibinfojournalEngineering Applications of Artificial Intelligence \bibinfovolume113, \bibinfopages104908. \bibinfopublisherElsevier.
  19. \bibinfotitleTime series data mining: A case study with big data analytics approach. \bibinfojournalIEEE Access \bibinfovolume8, \bibinfopages14322–14328. \bibinfopublisherIEEE.
  20. \bibinfoauthorHelmut Herwartz, \bibinfoyear2017. \bibinfotitleStock return prediction under GARCH—An empirical assessment. \bibinfojournalInternational Journal of Forecasting \bibinfovolume33(\bibinfonumber3), \bibinfopages569–580. \bibinfopublisherElsevier.
  21. \bibinfotitleStructured Ensembles: An approach to reduce the memory footprint of ensemble methods. \bibinfojournalNeural Networks \bibinfovolume144, \bibinfopages407–418. \bibinfopublisherElsevier.
  22. \bibinfotitleClip-q: Deep network compression learning by in-parallel pruning-quantization. \bibinfojournalProceedings of the IEEE conference on computer vision and pattern recognition \bibinfopages7873–7882.
  23. \bibinfotitleAn integrated model combined ARIMA, EMD with SVR for stock indices forecasting. \bibinfojournalInternational Journal on Artificial Intelligence Tools \bibinfovolume25, \bibinfopages1650005. \bibinfopublisherWorld Scientific.
  24. \bibinfotitleStock forecasting using local data. \bibinfojournalIEEE Access \bibinfovolume9, \bibinfopages9334–9344. \bibinfopublisherIEEE.
  25. \bibinfotitleFractional neuro-sequential ARFIMA-LSTM for financial market forecasting. \bibinfojournalIEEE Access \bibinfovolume8, \bibinfopages71326–71338. \bibinfopublisherIEEE.
  26. \bibinfotitleImproving forecasting accuracy of time series data using a new ARIMA-ANN hybrid method and empirical mode decomposition. \bibinfojournalNeurocomputing \bibinfovolume361, \bibinfopages151–163. \bibinfopublisherElsevier.
  27. \bibinfotitleA novel hybrid model based on recurrent neural networks for stock market timing. \bibinfojournalSoft Computing \bibinfovolume24, \bibinfopages15273–15290. \bibinfopublisherSpringer.
  28. \bibinfotitleStock closing price prediction based on sentiment analysis and LSTM. \bibinfojournalNeural Computing and Applications \bibinfovolume32, \bibinfopages9713–9729. \bibinfopublisherSpringer.
  29. \bibinfotitleA CNN-BiLSTM-AM method for stock price prediction. \bibinfojournalNeural Computing and Applications \bibinfovolume33, \bibinfopages4741–4753. \bibinfopublisherSpringer.
  30. \bibinfotitleChina’s commercial bank stock price prediction using a novel K-means-LSTM hybrid approach. \bibinfojournalExpert Systems with Applications \bibinfovolume202, \bibinfopages117370. \bibinfopublisherElsevier.
  31. \bibinfotitleMRC-LSTM: a hybrid approach of multi-scale residual CNN and LSTM to predict bitcoin price. In \bibinfobooktitle2021 International Joint Conference on Neural Networks (IJCNN), \bibinfopages1–8. \bibinfoorganizationIEEE.
  32. \bibinfotitleInformer: Beyond efficient transformer for long sequence time-series forecasting. \bibinfobooktitleProceedings of the AAAI conference on artificial intelligence, \bibinfovolume35, \bibinfonumber12, \bibinfopages11106–11115. \bibinfopublisherAAAI Press.
  33. \bibinfotitleTransformer-based attention network for stock movement prediction. \bibinfojournalExpert Systems with Applications \bibinfovolume202, \bibinfopages117239. \bibinfopublisherElsevier.
  34. Ha Young Kim, Chang Hyun Won, ”Forecasting the volatility of stock price index: A hybrid model integrating LSTM with multiple GARCH-type models”, Expert Systems with Applications, 103, 25–37, 2018. Elsevier.
  35. Heng-Li Yang, Han-Chou Lin, ”An integrated model combined ARIMA, EMD with SVR for stock indices forecasting”, International Journal on Artificial Intelligence Tools, 25(02), 1650005, 2016. World Scientific.
  36. ”A hybrid model for financial time-series forecasting based on mixed methodologies”, Expert Systems, 38(2), e12633, 2021. Wiley Online Library.
  37. ”Container throughput forecasting using a novel hybrid learning method with error correction strategy”, Knowledge-Based Systems, 182, 104853, 2019. Elsevier.
  38. ”Forecasting Day-Ahead Brent Crude Oil Prices Using Hybrid Combinations of Time Series Models”, Mathematics, 11(16), 3548, 2023. MDPI.
  39. \bibinfoauthorLiang-Ying Wei, \bibinfoyear2016. \bibinfotitleA hybrid ANFIS model based on empirical mode decomposition for stock time series forecasting. \bibinfojournalApplied Soft Computing, \bibinfovolume42, \bibinfopages368–376. \bibinfopublisherElsevier.
  40. \bibinfotitleForecasting selected colombian shares using a hybrid ARIMA-SVR model. \bibinfojournalMathematics, \bibinfovolume10(13), \bibinfopages2181. \bibinfopublisherMDPI.
  41. \bibinfotitleFractional frequency hybrid model based on EEMD for financial time series forecasting. \bibinfojournalCommunications in Nonlinear Science and Numerical Simulation, \bibinfovolume89, \bibinfopages105281. \bibinfopublisherElsevier.
  42. \bibinfotitleSoft computing model coupled with statistical models to estimate future of stock market. \bibinfojournalNeural Computing and Applications, \bibinfovolume33, \bibinfopages7629–7647. \bibinfopublisherSpringer.
  43. \bibinfotitlePredicting direction of stock price index movement using artificial neural networks and support vector machines: The sample of the Istanbul Stock Exchange. \bibinfojournalExpert systems with Applications \bibinfovolume38(5), \bibinfopages5311–5319. \bibinfopublisherElsevier.
  44. \bibinfotitleUsing artificial neural network models in stock market index prediction. \bibinfojournalExpert systems with Applications \bibinfovolume38(8), \bibinfopages10389–10397. \bibinfopublisherElsevier.
  45. \bibinfoauthorS.B. Achelis, \bibinfoyear2005. \bibinfotitleTechnical Analysis From A To Z. \bibinfopublisherVision Books. \bibinfourlhttps://books.google.com/books?id=J1YDPgAACAAJ
  46. \bibinfotitleVariational mode decomposition. \bibinfojournalIEEE transactions on signal processing, \bibinfovolume62(\bibinfonumber3), \bibinfopages531–544. \bibinfopublisherIEEE.
  47. \bibinfotitleUsing support vector regression and K-nearest neighbors for short-term traffic flow prediction based on maximal information coefficient. \bibinfojournalInformation Sciences, \bibinfovolume608, \bibinfopages517–531. \bibinfopublisherElsevier.
  48. \bibinfotitleMeasuring complexity using fuzzyen, apen, and sampen. \bibinfojournalMedical Engineering & Physics, \bibinfovolume31, \bibinfonumber1, \bibinfopages61–68. \bibinfopublisherElsevier.
  49. \bibinfotitleAttention is all you need. \bibinfojournalAdvances in neural information processing systems, \bibinfovolume30. \bibinfopublisherNIPS.
  50. \bibinfotitleStock price prediction using LSTM, RNN and CNN-sliding window model. \bibinfobooktitle2017 international conference on advances in computing, communications and informatics (icacci), \bibinfopages1643–1647. \bibinfopublisherIEEE.
  51. \bibinfotitleRiver water quality parameters prediction method based on LSTM-RNN model. \bibinfobooktitle2019 Chinese Control And Decision Conference (CCDC), \bibinfopages3024–3028. \bibinfopublisherIEEE.
  52. \bibinfotitleOn the difficulty of training recurrent neural networks. \bibinfobooktitleInternational conference on machine learning, \bibinfopages1310–1318. \bibinfopublisherPmlr.
  53. \bibinfoauthorJ. T. Barron, \bibinfoyear2019. \bibinfotitleA general and adaptive robust loss function. \bibinfobooktitleProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, \bibinfopages4331–4339. \bibinfopublisherIEEE/CVF.
  54. \bibinfotitleGradient centralization: A new optimization technique for deep neural networks. \bibinfojournalComputer Vision–ECCV 2020: 16th European Conference, Glasgow, UK, August 23–28, 2020, Proceedings, Part I 16 \bibinfovolume, \bibinfopages635–652. \bibinfopublisherSpringer.

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now