Ranking Momentum Signals for Entropy Pooling Portfolios

Summary

The document asks how to rank portfolio assets by momentum indicators for use as views in an Entropy Pooling framework. It describes Entropy Pooling as a way to blend views into a prior distribution, and relates the approach to Black–Litterman. Rate of Change is offered as a simple example: assets with stronger performance receive higher ranks.

The document provides no tested strategy, comparison of indicators, or evidence about performance. It is an open question rather than a guide to implementation, so it does not establish which indicators are suitable or how to handle lookback periods, ties, or different asset volatilities. Its useful contribution is the framing of ranked momentum signals as inputs to a portfolio distribution update.

Key ideas

• Entropy Pooling blends investment views into a reference distribution.
• The author proposes ranking portfolio assets by momentum indicators.
• Rate of Change is given as a basic example, with better performance receiving a higher rank.
• The document does not compare indicators or provide evidence about their effectiveness.

Tags

# Momentum strategy with Entropy Pooling


# Momentum strategy with Entropy Pooling












i'm currently trying to implement a ranking based on momentum indicators into my Entropy Pooling approach. Basically, the idea behind Entropy Pooling is to incorporate views into a reference model (prior distribution) by blending them. This procedure is a generalization and enhancement of the Black-Litterman model.

My Question is now: Do you have suggestions on momentum indicators, which are applicable on a ranking. By this i mean, that i want to rank the portfolio assets is an ascending order regarding the respective indicator. A first simple example could be ROC (Rate of Change), where the best performing asset is assigned the highest rank. Are there further indicators, which can be ranked in an easy way?

I would be very pleased on additional suggestions.

Thanks and regards!

Shown in full with attribution under the source's licence. Licence: CC BY-SA 4.0 (Stack Exchange)

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