Testing Four Spread Strategies on Moscow Exchange Futures

Summary

The article develops a spread between Si currency futures and RTS index futures, which it describes as usually moving in opposite directions. It fits a linear regression between the instruments, treats the difference from the resulting synthetic series as the spread, and visualizes it with indicators. Four entry approaches are explored: changes in the spread regression slope, spread sign changes on completed bars, sign changes confirmed over a run of ticks, and the spread reaching a threshold relative to the first instrument’s price.

The strategies are debugged and optimized in a multi-symbol tester using recorded tick data. The reported two-month historical exercise finds weak or inconsistent outcomes for the first three approaches, while one direction of the threshold strategy appears comparatively better. The article explicitly cautions that these limited-period optimization results may be random and should not guide live trading. It is primarily an example of building and testing hypotheses, not evidence of a robust spread-trading edge.

Key ideas

• The spread is calculated as the difference between one futures price and a regression-based synthetic value of the other.
• A regression slope, completed-bar sign change, tick-confirmed sign change, and percentage threshold are tested as entry signals.
• The tick-confirmation variant seeks to filter transient spread changes by requiring the sign to persist across several ticks.
• The reported optimization results are mixed, with the first three approaches performing poorly in the selected period.
• A short historical optimization is vulnerable to chance and does not establish a durable trading advantage.

Tags

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