Since the papers about realized volatility calculate daily volatility out of intraday data, is it also possible to apply same methods to calculate e.g. 10 minutes volatility by smaller sampled data e.g. by 1 second returns? I could not find a single paper about that. Specifically, I'm interest if methods which remove microstructure noise, such as kernels, are still valid.

  • $\begingroup$ Can you give the references of the papers you have in mind? You seem to think "kernel methods remove microstructure noise" $\endgroup$ – lehalle Jan 21 '17 at 8:16
  • $\begingroup$ They actually do, since the noise creates autocorrelation in the price process which is removed by the kernel. e.g. onlinelibrary.wiley.com/doi/10.3982/ECTA6495/abstract $\endgroup$ – Filippo Scopel Jan 22 '17 at 9:37

Your Answer

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

Browse other questions tagged or ask your own question.