Paper: arXiv 2511.07014

Authors: So-Yoon Cho, Jin-Young Kim, Kayoung Ban, Hyeng Keun Koo, Hyun-Gyoon Kim

Abstract

Probabilistic forecasting is crucial in multivariate financial time-series for constructing efficient portfolios that account for complex cross-sectional dependencies. In this paper, we propose Diffolio, a diffusion model designed for multivariate financial time-series forecasting and portfolio construction. Diffolio employs a denoising network with a hierarchical attention architecture, comprising both asset-level and market-level layers. Furthermore, to better reflect cross-sectional correlations, we introduce a correlation-guided regularizer informed by a stable estimate of the target correlation matrix. This structure effectively extracts salient features not only from historical returns but also from asset-specific and systematic covariates, significantly enhancing the performance of forecasts and portfolios. Experimental results on the daily excess returns of 12 industry portfolios show that Diffolio outperforms various probabilistic forecasting baselines in multivariate forecasting accuracy and portfolio performance. Moreover, in portfolio experiments, portfolios constructed from Diffolio’s forecasts show consistently robust performance, thereby outperforming those from benchmarks by achieving higher Sharpe ratios for the mean-variance tangency portfolio and higher certainty equivalents for the growth-optimal portfolio. These results demonstrate the superiority of our proposed Diffolio in terms of not only statistical accuracy but also economic significance.

Complexity vs Empirical Score

  • Math Complexity: 8.5/10
  • Empirical Rigor: 6.0/10
  • Quadrant: Holy Grail — high math complexity, high empirical rigor

Why this score: The paper employs advanced mathematics including diffusion models, hierarchical attention networks, and a correlation-guided regularizer with derivations. Empirical evaluation uses daily excess returns on 12 industry portfolios, reporting Sharpe ratios and certainty equivalents, but lacks code or detailed implementation details.

Research Flowchart

  flowchart TD
  A["Research Goal<br/>Multivariate Probabilistic Financial<br/>Time-Series Forecasting & Portfolio Construction"] --> B["Data & Inputs<br/>Historical Daily Excess Returns<br/>Asset & Market Covariates"]
  B --> C["Methodology: Diffolio Model<br/>Diffusion Process + Denoising<br/>Hierarchical Attention: Asset & Market Levels"]
  C --> D["Key Component<br/>Correlation-Guided Regularizer<br/>Using Stable Target Correlation Matrix"]
  D --> E["Computational Processes<br/>Forward/Reverse Diffusion<br/>Network Training & Inference"]
  E --> F["Key Outcomes<br/>1. Superior Multivariate Forecasting Accuracy<br/>2. Efficient Portfolio Construction"]
  F --> G["Final Results<br/>Higher Sharpe Ratios & Certainty Equivalents<br/>vs. Benchmarks"]

Cited References (59)

  1. DeepAR: Probabilistic forecasting with autoregressive recurrent networks (2020)
  2. Capital market equilibrium with transaction costs (1986)
  3. Liquidity premia and transaction costs (2007)
  4. Time-series generative adversarial networks (2019)
  5. AlphaPortfolio: Direct construction through deep reinforcement learning and interpretable AI (2021)
  6. Timevae: A variational auto-encoder for multivariate time series generation (2021) — link
  7. Sig-Wasserstein GANs for conditional time series generation (2024)
  8. Diffwave: A versatile diffusion model for audio synthesis (2020) — link
and 51 more
  1. Csdi: Conditional score-based diffusion models for probabilistic time series imputation (2021)
  2. Diffusion-based time series imputation and forecasting with structured state space models (2022) — link
  3. Autoregressive denoising diffusion models for multivariate probabilistic time series forecasting (2021)
  4. Predict, refine, synthesize: Self-guiding diffusion models for probabilistic time series forecasting (2023)
  5. Generative learning for financial time series with irregular and scale-invariant patterns (2024)
  6. Diffusion-ts: Interpretable diffusion for general time series generation (2024) — link
  7. Mg-tsd: Multi-granularity time series diffusion models with guided learning process (2024) — link
  8. Modeling financial time-series with generative adversarial networks (2019)
  9. Quant GANs: deep generation of financial time series (2020)
  10. Fin-gan: Forecasting and classifying financial time series via generative adversarial networks (2024)
  11. Factorvae: A probabilistic dynamic factor model based on variational autoencoder for predicting cross-sectional stock returns (2022)
  12. Generation of synthetic financial time series by diffusion models — → 2410.18897
  13. Wavenet: A generative model for raw audio (2016) — link
  14. A comprehensive look at the empirical performance of equity premium prediction (2008)
  15. Empirical asset pricing via machine learning (2020)
  16. Denoising diffusion probabilistic models (2020)
  17. Denoising diffusion implicit models (2020) — link
  18. Deep unsupervised learning using nonequilibrium thermodynamics (2015)
  19. Honey, I shrunk the sample covariance matrix (2003)
  20. Improved estimation of the covariance matrix of stock returns with an application to portfolio selection (2003)
  21. Hierarchical attention networks for document classification (2016)
  22. Characteristics are covariances: A unified model of risk and return (2019)
  23. Autoencoder asset pricing models (2021)
  24. Empirical properties of asset returns: stylized facts and statistical issues (2001)
  25. Portfolio Selection (1952)
  26. Probabilistic forecasting (2014)
  27. The dividend-price ratio and expectations of future dividends and discount factors (1988)
  28. Strictly proper scoring rules, prediction, and estimation (2007)
  29. Generative adversarial nets (2014)
  30. Auto-encoding variational bayes (2013) — link
  31. Flow matching for generative modeling (2023)
  32. Attention is all you need (2017)
  33. Layer normalization (2016) — link
  34. Perception encoder: The best visual embeddings are not at the output of the network (2025) — link
  35. Industry costs of equity (1997)
  36. Tree-based conditional portfolio sorts: The relation between past and future stock returns (2016)
  37. Deep learning in asset pricing (2024)
  38. Predicting residential energy consumption by explainable deep learning with long-term and short-term latent variables (2023)
  39. Explainable prediction of electric energy demand using a deep autoencoder with interpretable latent space (2021)
  40. Capital asset prices: A theory of market equilibrium under conditions of risk (1964)
  41. The arbitrage theory of capital asset pricing (1976)
  42. A financial time series denoiser based on diffusion models (2024)
  43. CoFinDiff: Controllable Financial Diffusion Model for Time Series Generation — → 2503.04164
  44. Estimation with quadratic loss (1961)
  45. Dynamic conditional correlation: A simple class of multivariate generalized autoregressive conditional heteroskedasticity models (2002)
  46. Betting against beta (2014)
  47. MarketGANs: Multivariate financial time-series data augmentation using generative adversarial networks — → 2601.17773
  48. Mgsfformer: A multi-granularity spatiotemporal fusion transformer for air quality prediction (2025)
  49. Ginar+: A robust end-to-end framework for multivariate time series forecasting with missing values (2025)
  50. Charging strategies optimization for lithium-ion battery: Heterogeneous ensemble surrogate model-assisted advanced multi-objective optimization algorithm (2025)
  51. MetaGNSDformer: Meta-learning enhanced gated non-stationary informer with frequency-aware attention for point-interval remaining useful life prediction of lithium-ion batteries (2026)

3 of 59 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2518 linked papers.