I would like to calculate intraday tick volatility between a time interval A to B.
E.g. If I have the quote and trade ticks for an instrument between a time interval e.g. 2023-02-17 10:00:02 to 2023-02-17 14:30:00, would the volatility during this interval be the SD of tick by tick returns? Or would it be better to resample the data to e.g. 1 min bins and then calculate the SD of these 1 min bin returns? Also, would it be better to use trade ticks, or the midprice of quote ticks? What is the correct way to annualize this calculated volatility (if required)?
Extending this, if I want to calculate the historical (e.g. 10 days) intraday tick volatility during this same period A->B, would the calculation be to average the individual intraday tick volatilities for each of the 10 days? E.g.
[math]\frac{1}{10} * (TickVol(A,B)_{t-1} + ... + TickVol(A,B)_{t-10})[/math]
I'm not sure if this method is right but it is the only way I can think of now.
E.g. If I have the quote and trade ticks for an instrument between a time interval e.g. 2023-02-17 10:00:02 to 2023-02-17 14:30:00, would the volatility during this interval be the SD of tick by tick returns? Or would it be better to resample the data to e.g. 1 min bins and then calculate the SD of these 1 min bin returns? Also, would it be better to use trade ticks, or the midprice of quote ticks? What is the correct way to annualize this calculated volatility (if required)?
Extending this, if I want to calculate the historical (e.g. 10 days) intraday tick volatility during this same period A->B, would the calculation be to average the individual intraday tick volatilities for each of the 10 days? E.g.
[math]\frac{1}{10} * (TickVol(A,B)_{t-1} + ... + TickVol(A,B)_{t-10})[/math]
I'm not sure if this method is right but it is the only way I can think of now.