Skip to content
All library documents

Using GARCH Estimates and Realized Volatility from High-Frequency Data

Article Quant Q&A · Author: Hans
Read the original

Summary

The document distinguishes estimating volatility with a GARCH model from forecasting future volatility. It also considers realized volatility calculated from high-frequency price observations as another way to quantify volatility. Its central point is that GARCH can produce volatility estimates, but using those estimates to estimate realized volatility is generally redundant because realized volatility is itself an estimate.

The answer allows that fitted GARCH volatility could technically be interpreted as an estimate of realized volatility, while emphasizing that this is not the usual purpose. No data, model fitting details, comparison, or empirical evidence is supplied. The note is therefore a conceptual distinction, and it does not explain how to construct realized volatility or specify when either measure is preferable in practice.

Key ideas

  • GARCH can be used to estimate volatility as well as forecast it.
  • High-frequency observations can be used to calculate realized volatility.
  • Modeling realized volatility with GARCH is generally redundant when realized volatility is already available.
  • The document gives a conceptual distinction without empirical comparison or implementation detail.

Tags

VolatilityStatisticsFutures
Full text
# Can one estimate rather than forecast volatility using the GARCH model?


# Can one estimate rather than forecast volatility using the GARCH model?












Can one use the GARCH model to estimate the realized variance/volatility, such as done in this paper, rather than forecast the volatility, from (high frequency) price/tick data?

## Answer by Richard Hardy (score 1)

https://quant.stackexchange.com/a/71161

You can use GARCH to estimate volatility. If you have high-frequency data, you can use realized volatility to estimate volatility. You would normally not use GARCH to estimate realized volatility (why estimate an estimate?), though I suppose it is technically possible to view your fitted volatilities from a GARCH model as estimates of realized volatilities.

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.