Skip to content
All library documents

Choosing Investor Views for Black–Litterman Portfolios

Article Quant Q&A · Author: Nourhaine Nefzi
Read the original

Summary

The document explains that Black–Litterman views can come from any researched source of expected excess return. Examples include published forecasts, broker recommendations, machine-learning models, technical analysis that produces target prices, and business-cycle indicators. Each candidate source should be evaluated for how it generates alpha before its predictions are used as portfolio views.

The response notes that views in the framework are generally modeled as normally distributed random variables, so the investor must make assumptions about their distributions. It gives examples of possible inputs but provides no empirical comparison, validation procedure, or guidance on estimating view uncertainty. The examples therefore suggest sources to investigate rather than establish that any one source is reliable or suitable for a particular portfolio.

Key ideas

  • Potential alpha sources can be translated into investor views for Black–Litterman optimization.
  • Candidate inputs include forecasts, broker recommendations, machine learning, technical analysis, and business-cycle indicators.
  • Research should assess whether a proposed source can produce alpha before using its predictions.
  • Views are modeled as normally distributed random variables, which requires assumptions about their distributions.

Tags

Portfolio constructionMachine learningTechnical indicatorsStatistics
Full text
# How to generate the views in Black-Litterman model?


# How to generate the views in Black-Litterman model?












I want to apply a Black-Litterman approach for portfolio optimization. My question is how to select investor views? I need to base the choice on a model. I would be thankful if you could give me some references or suggestions.

## Answer by Alexander Didenko (score 4, accepted)

https://quant.stackexchange.com/a/30372

Any potential source of "alpha" would suffice, in fact. And your research would be research of how this "alpha" source is able to produce alpha. On my mind, candidates could be (1) some well-documented predictions of somebody (like Prechter or Dow) - in that case you'll have 2-3 views for period, and the rest of assets or classes remain in equilibrium - see for example this paper by Batyrbekova - http://cyberleninka.ru/article/n/using-elliott-wave-theory-predictions-as-inputs-in-equilibrium-portfolio-models-with-views; (2) or broker recommendations (ANR in Bloomberg); (3) or some machine learning algorithm, like random forest - something similar to that http://cyberleninka.ru/article/n/application-of-ensemble-learning-for-views-generation-in-meucci-portfolio-optimization-framework; (4) or even traditional technical analysis, some method which would give you target prices; (5) see also this: http://cyberleninka.ru/article/n/cycle-adjusted-capital-market-expectations-under-black-litterman-framework-in-global-tactical-asset-allocation ; the author uses FED business cycle indicator.

Remember that in Black-Litterman view is normally distributed random variable, so you would have to make certain assumptions.

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.