Spread Trading and Scikit-Learn CTA Signal Mining

Summary

This event announcement outlines two educational topics: futures and cash-futures spread trading, and machine-learning methods for developing CTA signals. The spread section describes analyzing related contracts through their price difference, looking for suitable combinations, modeling spread time series, and building a spread grid strategy. It also names execution approaches and configuration features in VeighNa’s spread trading module.

The machine-learning section proposes using Scikit-Learn for feature analysis and signal research, including KBins clustering, feature-correlation heatmaps, vectorized performance analysis, and event-driven backtesting. It also mentions combining mean-reverting spread strategies with trend-following CTA strategies when constructing a portfolio. The document is an outline for an in-person event, not a full tutorial: it provides no strategy specifications, backtest results, data details, or evidence that the proposed approaches are profitable. Its claims about relative risk and performance are not supported with quantitative analysis.

Key ideas

• Spread trading focuses on the price difference between related futures contracts or cash and futures instruments.
• The proposed workflow includes selecting spread pairs, modeling their time series, and implementing a spread grid strategy.
• The event describes execution tools and configurable spread algorithms in VeighNa’s trading module.
• Scikit-Learn is presented as a toolkit for CTA signal research using feature analysis and clustering.
• The announcement suggests combining mean-reversion and trend-following strategies, but supplies no performance evidence.

Tags

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