Choosing Historical Volatility Measures for Fund Alpha Analysis

Summary

The document asks how to measure volatility over a past sample when the goal is explanatory analysis rather than forecasting. The motivating study relates monthly mutual fund alphas to market volatility measured over the same historical period. It contrasts a simple standard deviation with models such as GARCH and asks whether a more complex model is useful for this purpose.

The response recommends using realized historical volatility for an analysis of past relationships, while suggesting a forward-looking volatility index when the research question concerns near-future conditions. It also mentions a volatility-tracking fund as a possible proxy and cautions that model-based estimates introduce modeling error. The exchange offers advice rather than a comparison of estimators or empirical evidence, and it does not specify the return frequency, volatility calculation, or proxy selection criteria. Researchers would need to align the measure’s timing and construction with the hypothesis being tested.

Key ideas

• The choice of volatility measure should match whether the analysis concerns past or future conditions.
• Historical realized volatility is suggested for studying relationships within a past sample.
• A forward-looking volatility index may be relevant when the hypothesis concerns near-future conditions.
• Model-based estimates can introduce error, and the document does not compare alternatives empirically.

Tags

# In-sample volatility measurement


# In-sample volatility measurement












I would like to know what is the most reasonable way to measure volatility in a sample of past observations. Aside from standard deviation, are more complex models like GARCH used for (historical) volatility measurement if one is not interested in forecasting future volatility?

For context, as mentioned in a comment below, I need a measure of past monthly volatility to study the relationship between (monthly) mutual fund alphas and (monthly) market volatility over a past period of time.

## Answer by Dhruv Mahajan (score 0)

https://quant.stackexchange.com/a/45711

Why would you need to model volatility to test an hypothesis. Just use the historical realised volatility and if you want to test the hypothesis how funds relate in the near future, then use the VIX index, it's a forward looking measure. Or you use some volatility tracking fund as a proxy, why use some model to estimate relationships when obviously some modelling errors will creep in.

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.