Constraining Portfolio Risk While Optimizing Return

Summary

The document asks how to maximize expected portfolio return while keeping risk, such as value at risk, under a chosen ceiling. The answer points to an R package with portfolio optimization routines for expected shortfall and its long-short variant, suggesting these as ways to incorporate a risk target into allocation decisions. The question frames the problem across eight assets and contrasts a lower-risk allocation with a higher-risk one, but the reply does not show resulting weights or a worked comparison.

The recommendation is brief and provides no data, objective-function details, or instructions for setting the exact constraint. Expected shortfall and value at risk are related tail-risk measures, but they are not interchangeable; users should confirm what each function constrains and how the package defines its inputs. The document offers a possible implementation direction, not evidence that a particular allocation achieves the stated target or maximizes returns under realistic assumptions.

Key ideas

• The allocation problem is to maximize expected return subject to a specified risk ceiling.
• The reply recommends exploring R optimization routines for expected shortfall and long-short portfolios.
• The document provides no example weights, portfolio results, or comparison across risk targets.
• Users should check whether the package’s expected-shortfall objective matches the intended value-at-risk constraint.

Tags

# How to maintain VaR at 5% and max return in Portfolio Analytics


# How to maintain VaR at 5% and max return in Portfolio Analytics












For Asset Allocation in R using Portfolio Analytics, is there a way to set risk as constant number, then optimize portfolio returns? For example, to maintain VaR always at 5% (conservative), how do weights of 8 assets change in portfolio to max return? In contrast, how do the weights change compared to a risky (VaR =20%) portfolio? In the Portfolio Analytics package, we can only set min risk as objective, but not set risk as a constant number. (Different from Equal Risk Contribution)

## Answer by Gogo78 (score 1)

https://quant.stackexchange.com/a/49945

You can use the ‘portfolio.optimization' Package with this you can optimize your portfolio with a VaR set at 5% Max target and try to find the right allocation by using the function :

> optimal.portfolio.expected.shortfall

or you can optimize by CVaR:

> optimal.portfolio.expected.shortfall.long.short

... this is quite straight forward!!

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.