Using Continuous Signals to Analyze Crypto Trends

Summary

The document argues that binary rules, such as taking a position based only on whether price is above a moving average, discard information and conceal how signal strength relates to future returns. For a crypto trend example, it replaces the on/off indicator with the ratio of price to its moving average. This continuous feature assigns a value at each observation, allowing analysis of the full relationship rather than only two states.

The article describes plotting the signal against forward returns, grouping signal values into ten equal-sized buckets, and comparing cumulative returns attributed to the feature. The reported patterns are noisy and broadly increasing with signal strength, with a possible reversal among the most negative bucket. These views can suggest position scaling or further investigation, but the cumulative series is not a complete backtest: it excludes costs and real-world trading constraints. The observed effects are exploratory and do not establish that a deployable strategy will be profitable.

Key ideas

• Binary threshold signals hide variation in signal strength and discard observations’ information.
• The price-to-moving-average ratio provides a continuous measure of trend at each observation.
• Scatter plots and equal-sized signal buckets can reveal relationships and nonlinear patterns in forward returns.
• A noisy signal-return relationship may motivate scaling position size by feature magnitude, subject to further testing.
• Returns attributed to a signal are not a full backtest unless costs and trading constraints are included.

Tags

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