So-Yoon Cho, Jin-Young Kim, Kayoung Ban, Hyeng Keun Koo, Hyun-Gyoon Kim
arXiv 10 Nov 2025 · cs.CE · publishedInformation Fusion (2026)
arXiv:2511.07014 · PDF · DOI · OpenAlex · Extracted main text
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.
appendix boundary found by appendix_command · 77% of the source is main text. Read the extracted text to check this.
The works this paper leans on most, across its whole bibliography — not restricted to papers in our corpus. Ranked by composite intensity, which combines how often a work is mentioned, how many sections mention it, and how much of that falls in the main text rather than the appendix.
| Reference | Intensity | Mentions | Sections | Main text | |
|---|---|---|---|---|---|
| 1 | Gu, Shihao and Kelly, Bryan and Xiu, Dacheng (2020) Empirical asset pricing via machine learning | 0.974 | 13 | 4 | 92% |
| 2 | Ho, Jonathan and Jain, Ajay and Abbeel, Pieter (2020) Denoising diffusion probabilistic models | 0.941 | 6 | 5 | 83% |
| 3 | Welch, Ivo and Goyal, Amit (2008) A comprehensive look at the empirical performance of equity premium prediction | 0.941 | 6 | 4 | 83% |
| 4 | Fan, Xinyao and Wu, Yueying and Xu, Chang and Huang, Yuhao and Liu,… (2024) Mg-tsd: Multi-granularity time series diffusion models with guided learning process | 0.843 | 3 | 3 | 100% |
| 5 | Sohl-Dickstein, Jascha and Weiss, Eric and Maheswaranathan, Niru and… (2015) Deep unsupervised learning using nonequilibrium thermodynamics | 0.843 | 3 | 3 | 100% |
| 6 | Tashiro, Yusuke and Song, Jiaming and Song, Yang and Ermon, Stefano (2021) Csdi: Conditional score-based diffusion models for probabilistic time series imputation | 0.843 | 3 | 3 | 100% |
| 7 | Kelly, Bryan T and Pruitt, Seth and Su, Yinan (2019) Characteristics are covariances: A unified model of risk and return | 0.737 | 3 | 2 | 100% |
| 8 | Tanaka, Yuki and Hashimoto, Ryuji and Takayanagi, Takehiro and Piao,… (2025) CoFinDiff: Controllable Financial Diffusion Model for Time Series Generation | 0.737 | 3 | 2 | 100% |
| 9 | Alcaraz, Juan Miguel Lopez and Strodthoff, Nils (2022) Diffusion-based time series imputation and forecasting with structured state space models | 0.644 | 2 | 2 | 100% |
| 10 | Gneiting, Tilmann and Katzfuss, Matthias (2014) Probabilistic forecasting | 0.644 | 2 | 2 | 100% |
Showing the top 10 of 50 scored citations.
arXiv econ.EM papers that cite this one, ranked by how heavily they lean on it.
| Citing paper | Intensity | Mentions | Sections | |
|---|---|---|---|---|
| 1 | MarketGANs: Multivariate financial time-series data augmentation using generative adversarial networks | 0.511 | 2 | 1 |