Setting Tactical Asset Allocation Ranges from Risk and Information Ratios

Summary

The document asks whether tactical asset allocation ranges around strategic benchmark weights can be derived from asset-class correlations and information ratios without starting from proposed tactical weights or signals. It outlines two existing approaches: one uses initial tactical weights, a trade matrix, asset-class correlations, and information ratios to determine aggressiveness factors; another applies a similar idea to initial signals, correlations, and alphas. In both cases, the factors scale the starting tactical positions or signals to create ranges.

The accepted response points to David E. Kuenzi’s work on tracking error and tactical ranges as an approach that meets the stated requirement. The excerpt does not explain that method’s equations, assumptions, implementation, or empirical performance, so it identifies a reference rather than providing a usable procedure. It also does not specify how to estimate correlations or information ratios, or how the resulting ranges should be constrained in a live portfolio. Further reading is needed before applying the cited approach.

Key ideas

• Tactical ranges can be framed as adjustments around strategic asset allocation weights.
• The approaches described use correlation structure and information ratios or alphas to scale existing tactical weights or signals.
• The question seeks a method that does not require initial tactical weights or signals.
• The answer cites a tracking-error-based reference, but the excerpt does not provide its mechanics or evidence.

Tags

# Trading Ranges for Tactical Asset Allocation


# Trading Ranges for Tactical Asset Allocation












Do methods exist to determine trading ranges around benchmark weights/strategic asset allocation weights for a tactical asset allocation from the correlation structure between the individual asset classes and their respective information ratios?

I am aware of the following approaches:

- The approach laid out in Portfolio Construction and Risk Budgeting, Chapter 15, by Bernd Scherer. It takes an initial set of tactical asset allocation weights, the correlation structure derived from a trade matrix and the underlying asset class correlation structure and the information ratios of the asset classes and determines aggressiveness factors which are multiplied with initial tactical asset allocation weights to determine the corresponding ranges.

- The approach laid out in Advanced Theory and Methodology of Tactical Asset Allocation, Chapter 6, by Wai Lee. It takes an initial set of signals, the correlation structure of the asset classes and the alphas to determine aggressiveness factor which are applied to the initial set of signals to determine the corresponding ranges.

Are there approaches which are independent of an initial set of tactical weights/signals, but take the correlation structure between the asset classes and their information ratios into account?

## Answer by Hans-Peter Schrei (score 1, accepted)

https://quant.stackexchange.com/a/53560

An approach which satisfies the requirements I listed above is the one laid out in Tracking Error and the Setting of Tactical Ranges, David E. Kuenzi, The Journal of Investing Spring 2004, 13 (1) 35-44.

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.