Using Partial and Linear Orders to Select Trailing-Stop Steps

Summary

This article applies order theory to selecting steps in a trailing-stop process. It distinguishes partial orders, which allow some elements to remain incomparable, from linear orders, which require every element to be comparable. The author argues that partial ordering can retain ambiguous price patterns, such as doji candles, while linear ordering can support consistent ranking when comparing assets for portfolio decisions. In the proposed trading application, price-action autocorrelation indices determine the ordering of lookback, applied-price, and indicator sets, which in turn selects trailing-stop decisions.

The article reports a case study comparing partial-order and linear-order trailing stops. It says the partial-order version produced a more promising result, while the linear-order version had lower profitability and worse drawdown despite fewer trades. However, the author explicitly calls for longer tests across multiple symbols before drawing firm conclusions. The evidence is therefore preliminary and tied to the described system; it does not show that category-theory-based ordering generally improves trade exits or that the approach transfers to other strategies.

Key ideas

• Partial orders permit incomparable cases, while linear orders require every item to be ranked against every other item.
• The proposed method uses autocorrelation-based price patterns to select trailing-stop decision steps.
• The article argues that partial ordering can preserve ambiguous price-bar patterns in classifications.
• Its case study favors partial ordering, but the author says longer tests across multiple symbols are needed.
• The reported comparison does not establish general performance benefits for other trading systems.

Tags

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