Methods for Estimating Volatility from Intraday Price Data
Summary
The document considers how to estimate changing volatility from minute-level stock observations. It outlines simple proxies, including the range within each interval and the standard deviation of short-interval price changes, alongside conditional heteroskedasticity models such as the GARCH family. The more complex models can represent volatility clustering and support forecasts.
A second answer lists broader metrics, including extreme returns, standard deviation, counts of days exceeding a return threshold, and average true range. These measures capture different aspects of price movement and operate over different sampling windows. The discussion offers no empirical comparison on the reader’s data and does not specify details such as return construction, intraday seasonality adjustments, or handling market microstructure noise. It therefore presents alternatives to evaluate for a particular purpose rather than establishing one universally best estimator.
Key ideas
- Intraday range and return dispersion provide simple proxies for volatility.
- GARCH models can represent volatility clustering and produce forecasts.
- Extreme returns, threshold counts, and average true range measure distinct aspects of price variation.
- Estimator choice depends on the data and intended use, and should be compared empirically.
Tags
Full text
# How to calculate volatility on intraday data? # How to calculate volatility on intraday data? I have several weeks of minute-by-minute stock data (start and end prices, volume). Everything I've read so far leads me to believe there isn't a standard method for volatility, which is leaving me confused on what to use. I want to calculate the volatility for each minute so I can see how volatility is changing through the day and whether there is a correlation with other variables. What methods would apply here? ## Answer by Robert Szóstakowski (score 8, accepted) https://quant.stackexchange.com/a/20662 Financial modeling is often considered as a mixture of art and science. That is a way how you model your data depends on you. You can choose several approaches, for example: - calculate max - min price for a given minute data - a very simple approach, - calculate the standard deviation of minute-by-minute stock data, - calculate GARCH family models for measuring the volatility. This is a more complicated approach but gives you the most adapting models which can detect volatility clustering and thanks to it you can make volatility forecasts. You do not have to implement these models. They are implemented in a dedicated software (ox-metrics, gretl) or implemented in some open source libraries. To sum up. There is no final answer. Try to use different approaches and decide which suits best for you. Probably in the end of the day GARCH family models will appear the best. For more information, see: - https://en.wikipedia.org/wiki/Autoregressive_conditional_heteroskedasticity ## Answer by Koray Güclü (score 1) https://quant.stackexchange.com/a/21055 There are many ways to calculate the volatility. timeframe does not metter. it can be monthly quarterly or daily data. You can call them as volatility metrics. ## Volatility Metrics Volatility is the degree of trading price over a specific time window. Historical volatility is the degree of price changes of past market prices.Volatility indicates the risk your are taking by investing into a specific instrument. here are some simple methods ### Best and worst returns A volatility measure is to find worst and best returns for a specific time. For example, SPY has worst return -8% and best return +4%. Volatility measure is +4%/-8%. ### Standard Deviation Another measure is standard deviation of monthly returns. ## Number of trading days in access of 1% A fourth volatility metric is the number of trading days during a year on which the price of an asset increased or decreased by more than 1 percent. ### True Range Average true range (ATR) is a technical analysis volatility indicator originally developed by J. Welles Wilder, Jr. for commodities see : Average true range - Wikipedia, the free encyclopedia ### Standard Deviation ... I hope this helps
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.