Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
156 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

Deep Generative Modeling for Financial Time Series with Application in VaR: A Comparative Review (2401.10370v1)

Published 18 Jan 2024 in q-fin.CP, cs.LG, q-fin.RM, and q-fin.ST

Abstract: In the financial services industry, forecasting the risk factor distribution conditional on the history and the current market environment is the key to market risk modeling in general and value at risk (VaR) model in particular. As one of the most widely adopted VaR models in commercial banks, Historical simulation (HS) uses the empirical distribution of daily returns in a historical window as the forecast distribution of risk factor returns in the next day. The objectives for financial time series generation are to generate synthetic data paths with good variety, and similar distribution and dynamics to the original historical data. In this paper, we apply multiple existing deep generative methods (e.g., CGAN, CWGAN, Diffusion, and Signature WGAN) for conditional time series generation, and propose and test two new methods for conditional multi-step time series generation, namely Encoder-Decoder CGAN and Conditional TimeVAE. Furthermore, we introduce a comprehensive framework with a set of KPIs to measure the quality of the generated time series for financial modeling. The KPIs cover distribution distance, autocorrelation and backtesting. All models (HS, parametric and neural networks) are tested on both historical USD yield curve data and additional data simulated from GARCH and CIR processes. The study shows that top performing models are HS, GARCH and CWGAN models. Future research directions in this area are also discussed.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (41)
  1. Wasserstein gan. https://arxiv.org/abs/1701.07875.
  2. Non-parametric var techniques. myths and realities. Economic Notes by Banca Monte dei Paschi di Siena SpA 30(2), 167–181.
  3. Var without correlations for portfolios of derivatives securities. Journal of Futures Markets 19, 583–602.
  4. Basel Committee on Banking Supervision (2019). Minimum capital requirements for market risk. Bank for international settlements.
  5. Berkowitz, J. (2001). Testing density forecasts, with applications to risk management. Journal of Business & Economic Statistics 19(4), 465–474.
  6. Bollerslev, T. (1986). Generalized Autoregressive Conditional Heteroskedasticity. Journal of Economics 31(3), 307–327.
  7. Generative adversarial networks in time series: A survey and taxonomy. https://arxiv.org/abs/2107.11098.
  8. Generative adversarial networks in time series: A systematic literature review. ACM Comput. Surv. 55(10). https://doi.org/10.1145/3559540.
  9. A primer on the signature method in machine learning. arXiv preprint arXiv:1603.03788.
  10. Tail-gan: Learning to simulate tail risk scenarios.
  11. A Theory of the Term Structure of Interest Rates. Econometrica 53(2), 385–407.
  12. Quality control. Risk 9(9), 138–143.
  13. Timevae: A variational auto-encoder for multivariate time series generation. https://arxiv.org/abs/2111.08095.
  14. Diffusion models beat gans on image synthesis. In M. Ranzato, A. Beygelzimer, Y. Dauphin, P. Liang, and J. W. Vaughan (Eds.), Advances in Neural Information Processing Systems, Volume 34, pp.  8780–8794. Curran Associates, Inc. https://proceedings.neurips.cc/paper/2021/file/49ad23d1ec9fa4bd8d77d02681df5cfa-Paper.pdf.
  15. Distribution of the estimators for autoregressive time series with a unit root. Journal of the American Statistical Association 74(366), 427–431.
  16. Forecasting the Term Structure of Government Bond Yields. Journal of Economics 130(2), 337–364.
  17. Dragulescu, A. A. and V. M. Yakovenko (2002). Probability distribution of returns in the heston model with stochastic volatility. https://arxiv.org/abs/cond-mat/0203046.
  18. Engle, R. F. (1982). "autoregressive conditional heteroskedasticity with estimates of the variance of United Kingdom inflation". Econometrica 50(4), 987–1007.
  19. Fu, R. (2022). An introduction of encoder-and-decoder cgan framework.
  20. Time series simulation by conditional generative adversarial net. https://arxiv.org/abs/1904.11419.
  21. Graves, A. (2013). Generating sequences with recurrent neural networks. https://arxiv.org/abs/1308.0850.
  22. Denoising diffusion probabilistic models. In H. Larochelle, M. Ranzato, R. Hadsell, M. Balcan, and H. Lin (Eds.), Advances in Neural Information Processing Systems, Volume 33, pp.  6840–6851. Curran Associates, Inc. https://proceedings.neurips.cc/paper/2020/file/4c5bcfec8584af0d967f1ab10179ca4b-Paper.pdf.
  23. Johnson, S. (2020). Synthetic data for finance - from theory to practice. https://www.ubs.com/global/en/investment-bank/in-focus/research-focus/quant-answers/quant-insight-series/_jcr_content/mainpar/toplevelgrid_7262680_322968126/col3/actionbutton_3358030.0949818704.file/PS9jb250ZW50L2RhbS9pbnRlcm5ldGhvc3RpbmcvaW52ZXN0bWVudGJhbmsvZW4vZXF1aXRpZXMvcWlzLXZpcnR1YWwtZXZlbnQtZGVja3MtMjAyMC9zeW50aGV0aWMtZGF0YS1mb3ItZmluYW5jZS1zdGVmYW4tamFuc2VuLXVicy0yMDIxLnBkZg==/synthetic-data-for-finance-stefan-jansen-ubs-2021.pdf.
  24. An introduction to variational autoencoders. Foundations and Trends® in Machine Learning 12(4), 307–392. https://doi.org/10.1561%2F2200000056.
  25. Foundations of sequence-to-sequence modeling for time series. https://arxiv.org/abs/1805.03714.
  26. Umap: Uniform manifold approximation and projection for dimension reduction.
  27. A generalised signature method for multivariate time series feature extraction. arXiv preprint arXiv:2006.00873.
  28. Nelson, C. R. and A. F. Siegel (1987). Parsimonious modeling of yield curve. Journal of Business 60, 473–489.
  29. Perignon, C. and D. R. Smith (2010). The level and quality of value-at-risk disclosure by commercial banks. Journal of Banking & Finance 34(2), 362–377.
  30. On wasserstein two sample testing and related families of nonparametric tests. https://arxiv.org/abs/1509.02237.
  31. Ramya Malur Srinivasan, A. C. (2020). Generating user-friendly explanations for loan denials using generative adversarial networks. https://www.risk.net/investing/quant-investing/7833326/in-fake-data-quants-see-a-fix-for-backtesting.
  32. Autoregressive denoising diffusion models for multivariate probabilistic time series forecasting. https://arxiv.org/abs/2101.12072.
  33. Sabaté, M. (2021). The authors’ official pytorch sigcwgan implementation.
  34. Denoising diffusion implicit models. https://arxiv.org/abs/2010.02502.
  35. Srinivasan, P. and W. J. Knottenbelt (2022). Time-series transformer generative adversarial networks. https://arxiv.org/abs/2205.11164.
  36. Sequence to sequence learning with neural networks. In Z. Ghahramani, M. Welling, C. Cortes, N. Lawrence, and K. Weinberger (Eds.), Advances in Neural Information Processing Systems, Volume 27. Curran Associates, Inc. https://proceedings.neurips.cc/paper/2014/file/a14ac55a4f27472c5d894ec1c3c743d2-Paper.pdf.
  37. van den Berg, T. (2011). Calibrating the ornstein-uhlenbeck (Vasicek) model. https://www.statisticshowto.com/wp-content/uploads/2016/01/Calibrating-the-Ornstein.pdf.
  38. van der Maaten, L. and G. Hinton (2008). Visualizing data using t-sne. Journal of Machine Learning Research 9(86), 2579–2605.
  39. Vasicek, O. (1977). An equilibrium characterization of the term structure. Journal of Financial Economics 5(2), 177–188.
  40. Villani, C. (2008). Optimal Transport: Old and New. Grundlehren der mathematischen Wissenschaften. Springer Berlin Heidelberg.
  41. Conditional LSTM-GAN for melody generation from lyrics. ACM Transactions on Multimedia Computing, Communications, and Applications 17(1), 1–20. https://doi.org/10.1145%2F3424116.
Citations (3)
View on Semantic Scholar

Summary

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

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