Research map / Time series / Foundation models for time series
CTBench: Cryptocurrency Time Series Generation Benchmark
Yihao Ang, Qiang Wang, Qiang Huang, Yifan Bao, Xinyu Xi, Anthony Kum Hoe Tung, Chen Jin, Zhiyong Huang
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use Diffusion-TS as the TSG model | Predictive Utility task · Predictive Utility, 2021 bull market | forecasting accuracy vs CAGR | Mixed |
| use TimeVAE as the TSG model | Predictive Utility task · Predictive Utility, 2021-2024 | Sharpe ratio / CAGR | Improved |
| use COSCI-GAN as the TSG model | Predictive Utility task · Predictive Utility, trend strategies (LOTQ, PW) | CAGR / Sharpe | Improved |
| use COSCI-GAN as the TSG model | Predictive Utility task · Predictive Utility, 2023 consolidation | trading outcomes | Hurt |
| use Fourier-Flow as the TSG model | Predictive Utility task · Predictive Utility, 2023 | risk-adjusted return | Improved |
| use LS4 as the TSG model | Predictive Utility task · Predictive Utility equity curves | returns | No clear effect |
| use KoVAE or LS4 as the TSG model | Statistical Arbitrage task · Statistical Arbitrage, 2022-2023 | CAGR | Improved |
| use FIDE as the TSG model | Statistical Arbitrage task · Statistical Arbitrage | CAGR / Sharpe | Hurt |
| introduce 0.03% trading fees | Statistical Arbitrage task · Statistical Arbitrage | profitability | Hurt |
| use smoother residuals (TimeVAE, Diffusion-TS) instead of high-turnover KoVAE under fees | Statistical Arbitrage task · Statistical Arbitrage with fees | Sharpe rank | Improved |
| lower reconstruction or prediction error of the TSG model | Predictive Utility task · Predictive Utility | trading success | No clear effect |
| use diffusion-based TSG models | TSG model zoo · Efficiency | training/inference time | Hurt |
Closest papers by meaning
- EarnHFT: Efficient Hierarchical Reinforcement Learning for High Frequency Trading (AAAI 2024 · similarity 0.83)
- MarS: a Financial Market Simulation Engine Powered by Generative Foundation Model (ICLR 2025 · similarity 0.83)
- The Temporal Graph of Bitcoin Transactions (NeurIPS 2025 · similarity 0.83)
- FinBen: A Holistic Financial Benchmark for Large Language Models (NeurIPS 2024 · similarity 0.82)
- LOB-Bench: Benchmarking Generative AI for Finance - an Application to Limit Order Book Data (ICML 2025 · similarity 0.82)
- TSGM: A Flexible Framework for Generative Modeling of Synthetic Time Series (NeurIPS 2024 · similarity 0.82)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
