Standardized Log Returns for SZI Mean-Reversion Signals

Summary

This strategy standardizes logarithmic returns using a rolling mean and standard deviation, calling the resulting z-score the Shiryaev-Zhou Index (SZI). It opens a long position when the score falls below a negative threshold and a short position when it rises above a positive threshold, treating unusually large moves as potential reversals. The described settings use a 50-period window and thresholds of two standard deviations, with percentage-based stop-loss and take-profit levels.

The document provides a daily BTC_USDT futures backtest configuration spanning several years, but gives no return, drawdown, or trade statistics. Its statistical rationale is presented as a premise rather than evidence that the signal is profitable. The text flags sensitivity to window and threshold choices, false signals in trending markets, execution slippage during volatile periods, and signal lag. The source also plots signals with a one-bar offset, so the relationship between displayed markers and executable timing merits scrutiny.

Key ideas

• The SZI is calculated by subtracting rolling average log return from current log return and dividing by rolling standard deviation.
• Extreme negative and positive SZI readings trigger potential long and short mean-reversion entries.
• The described setup uses a rolling window and fixed percentage stop-loss and take-profit levels.
• Trending markets, parameter sensitivity, slippage, and calculation delay are material limitations.
• The published backtest settings are not accompanied by performance results.

Tags

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