本文探讨了传统回测技术的局限性，介绍了合成人工数据在金融领域的应用，如创建平行市场数据来测试投资策略。通过使用神经网络生成的数据，量化投资者能够进行前向测试并应对市场未知风险。数据生成器可以复制真实市场的复杂模式，增强模拟的准确性，帮助揭示潜在风险。研究还显示，结合机器学习模型，如GAN和变分自编码器，可以提高回测的逼真度和效果。
文章：Rob Mannix, In fake data, quants see a fix for backtesting, Risk.net, May 2021.
传统回测技术
- One version of events: traditional backtesting uses only one version of history: investors test the effectiveness of a strategy by charting how it would have performed in real-world conditions – they see what already happened.
- Traditional methods of backtesting strategies is backward-looking.
- ‘New flavour’ to backtesting: make it possible to explore the market’s unknown unknowns.
合成人工数据
- Ultra-realistic ‘deep fake’ videos are common on social media platforms: videos of Tom Cruise playing golf and doing magic tricks raked up millions of views on TikTok. The clips didn’t feature Tom Cruise at all.
- The technology behind these ultra-realistic ‘deep fake’ videos has already been applied in finance: quants use it to create parallel universes of data to test investment strategies.
- Synthetic/artificial data: needs to invent alternative histories for deeper testing
- New models allow forward-looking testing: quants can input into a model the level of market volatility or features of recent price moves and ask the model to plot future paths based on those readings
- Rob Carver: “synthetic data could help in stress-testing strategies for which losses can be highly path dependent”
- Blanka Hor
最低0.47元/天 解锁文章
扫一扫
被折叠的 条评论
为什么被折叠?
到【灌水乐园】发言
