Choosing GARCH Variants for Asymmetric Volatility

Summary

The document considers whether there is a systematic way to choose among GARCH-family volatility models. The responses offer a practical modeling principle: match the model’s features to observed behavior in the market rather than selecting variants by trial and error. In particular, some returns may exhibit a leverage effect, where volatility increases more after price declines than after gains. The answer gives stock indexes as an example and points to GJR-GARCH as a model that incorporates this asymmetry.

The discussion also cautions that many ARCH-style extensions have been proposed without a clear justification, and that there is no universal ranking or selection system in the excerpt. It recommends familiarity with model use cases and informed judgment. No data, formal model comparison, or validation procedure is presented, so the guidance is conceptual: identify relevant return characteristics, choose a model capable of representing them, and assess its suitability for the intended application.

Key ideas

• GARCH variants should be selected with reference to market return behavior.
• Some markets exhibit a leverage effect, in which volatility responds more strongly to declines.
• GJR-GARCH is identified as one model that can represent this asymmetric response.
• The responses describe no universal systematic ranking of GARCH variants.
• Model choice requires understanding use cases and applying judgment; no empirical comparison is supplied.

Tags

# Uses of Volatility models


# Uses of Volatility models












I'm reading about volatility analysis here http://vlab.stern.nyu.edu/doc?topic=mdls. There are many variations of GARCH.

My question is: rather than trial-and-error approach, is there any systematic approach to decide when to use which type of GARCH? Or equivalently, each variation of GARCH is used in which situation?

## Answer by nbbo2 (score 2)

https://quant.stackexchange.com/a/18326

My 2 Zimbabwe cents:

A few years ago developing new ARCH like models became almost a fad and large numbers of them were published without a clear justification in my humble opinion.

However there is an important distinction I do think. Some markets are symmetric, while others (such as Stock Indexes) show a Leverage Effect where the volatility rises when the underlying goes down. It is important to use the right kind of model accordingly. For example for Stock Indexes I like GJR-GARCH which incorporates a leverage effect.

## Answer by Thering (score 1)

https://quant.stackexchange.com/a/18322

You won't find a systematic approach, that would require the models to be arranged in some system, whereas generally each is an adjustment of GARCH. The best you can do is to know the usage cases of as many as possible and then use your own judgement as to which model is appropriate.

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.