I would like to calculate realized volatility for a 15 min period. Most of the literature I looked up shows how to construct daily realized volatility using intraday data. These literatures does use several techniques (e.g. subsampling, using realised kernel...) to account for microstructure noise. I'm not sure if using these techniques for the 15 min period would help to reduce the microstructure noise since I use seconds data. Do you have any suggestions how to calculate realized volatility for a 15 min interval?

  • $\begingroup$ Very nice question, on a very interesting topic. I don't now about a precise method to get intraday (15 minutes, in this case) Realized Volatility measures, but, personally, I think that existent approaches to calculate daily volatilities can be applied to an intraday basis (such as the Two Scales Realized Volatility, wwwf.imperial.ac.uk/~pavl/AitSahalia2005.pdf . Can give a look to aabri.com/manuscripts/11881.pdf , to a comparison between RV and TSRV). This is just an opinion, I've no empirical confirmation. $\endgroup$ – simmy Jun 1 '16 at 16:04
  • $\begingroup$ FYI. I used now the approach in jfec.oxfordjournals.org/content/10/1/54.full.pdf. Its pretty simple and there is a R implementation. $\endgroup$ – Filippo Scopel Jun 7 '16 at 16:31

I can think two techniques that may possible be of help. The first technique is the moving average adjusted returns originally proposed by Andersen et. al(2001). See Hansen et al. (2008) for details. In order to account for serial correlation an MA(1) is fitted to to the intraday returns data, the residual of which is then squared and aggregated over the trading day to produce the daily measure of realised volatility. Besides, is the Hansen and Lunde first order autocorrelation realized volatility measure on the paper Realized Variance and Microstructure Noise.

| improve this answer | |

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.