Skip to content
All library documents

Benchmarking Synthetic Time Series for Cryptocurrency Trading

Article arXiv papers · Author: Yihao Ang et al.
Read the original

Summary

CTBench is a benchmark for evaluating synthetic cryptocurrency time series, motivated by the market’s continuous trading, high volatility, and rapid regime changes. It assembles data from 452 tokens and evaluates eight representative generation models across four market regimes. Its 13 metrics cover forecasting accuracy, cross-sectional ranking, trading outcomes, risk, and computational cost.

The benchmark uses two tasks: predictive utility, which tests whether generated data retains patterns useful for forecasting, and statistical arbitrage, which checks whether reconstructed series can support mean-reverting trading signals. The reported analysis finds trade-offs between statistical fidelity and profitability, and offers comparative rankings and model-selection guidance. The benchmark provides a framework for evaluating synthetic data, but its findings depend on the chosen models, metrics, market regimes, and the strategies used in evaluation; strong synthetic-data resemblance alone does not establish live trading value.

Key ideas

  • CTBench evaluates generated crypto time series for both predictive and trading uses.
  • Its metrics cover forecasting, rank fidelity, trading performance, risk, and computational efficiency.
  • The predictive utility task tests whether synthetic data preserves patterns useful for prediction.
  • The statistical arbitrage task evaluates support for mean-reverting trading signals.
  • The reported comparisons show that statistical fidelity and profitability can diverge.

Tags

CryptoMachine learningStatisticsBacktestingMean reversion
Full text
# CTBench: Cryptocurrency Time Series Generation Benchmark


# CTBench: Cryptocurrency Time Series Generation Benchmark









Synthetic time series are essential tools for data augmentation, stress testing, and algorithmic prototyping in quantitative finance. However, in cryptocurrency markets, characterized by 24/7 trading, extreme volatility, and rapid regime shifts, existing Time Series Generation (TSG) methods and benchmarks often fall short, jeopardizing practical utility. Most prior work (1) targets non-financial or traditional financial domains, (2) focuses narrowly on classification and forecasting while neglecting crypto-specific complexities, and (3) lacks critical financial evaluations, particularly for trading applications. To address these gaps, we introduce \textsf{CTBench}, the first comprehensive TSG benchmark tailored for the cryptocurrency domain. \textsf{CTBench} curates an open-source dataset from 452 tokens and evaluates TSG models across 13 metrics spanning 5 key dimensions: forecasting accuracy, rank fidelity, trading performance, risk assessment, and computational efficiency. A key innovation is a dual-task evaluation framework: (1) the \emph{Predictive Utility} task measures how well synthetic data preserves temporal and cross-sectional patterns for forecasting, while (2) the \emph{Statistical Arbitrage} task assesses whether reconstructed series support mean-reverting signals for trading. We benchmark eight representative models from five methodological families over four distinct market regimes, uncovering trade-offs between statistical fidelity and real-world profitability. Notably, \textsf{CTBench} offers model ranking analysis and actionable guidance for selecting and deploying TSG models in crypto analytics and strategy development.

Shown in full with attribution under the source's licence. Licence: abstract CC0

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.