
Consider strategy A that is exposed/"onrisk" with one specific position on average for 1 minute. A may exit the position after 1 minute, and immediately reenter another position for 1 minute, or it may pause to reenter the next position after X minutes. To gauge the volatility of the underlying such that the volatility informs A on the risk of the next position, is it appropriate to calculate the volatility of the asset over 1 minute intervals?
Furthermore and more broadly, what are the questions to consider when choosing the timeframe on which to calculate the volatility in this case? 
Nous promettons selon nos espérances, et nous tenons selon nos craintes. 


ronin


Total Posts: 591 
Joined: May 2006 


You need to rephrase your thinking.
Volatility doesn't mean anything. It's just a convenience number.
Presumably the question you are asking "what is the expected standard deviation of my position", or something like that. And, once you phrase it like that, it is kind of obvious what you should be calculating.
One glaring issue in this situation would be intraday seasonality. First minute after the open is very different from some minute during mid day when everybody is out for lunch. 
"There is a SIX am?"  Arthur 


Sure I meant standard deviation. I figure you mean I should be looking at 1 minute standard deviation, I don't find that obvious but then again I'm no expert. Noted for intraday seasonality 
Nous promettons selon nos espérances, et nous tenons selon nos craintes. 


ronin


Total Posts: 591 
Joined: May 2006 


Well, you should be looking at the standard deviation from the time you put the position on until you take it off.
It is simple as that.
Thinking in terms of volatility doesn't help, quite the opposite. 
"There is a SIX am?"  Arthur 


I'll take a simple answer wherever I can find one :) 
Nous promettons selon nos espérances, et nous tenons selon nos craintes. 



First you need to be specific on volatility.
Nowadays when one blurts the word vol you need to distinguish which exact vol?
Forward vol? Static vol? Implied? Realized?
Once that’s clairified then that type of volatility’s time frame will be determined by the traders horizon for opportunity.
Let’s say I’m trading ATM butterflies (naturals and skewed) in the $RUT within the implied space. My model gives me signals for trades with expirations <30DTE.
Thus I position my wings in accordance with volatility levels and some more metrics I’ll speak about later. The wings value is positioned relative to the short guts (ATMF straddle/MMM). This is if I’m debating whether to use an iron or a put/call, sometimes this is based off microstructure. Since gamma curvature is ON MY SIDE, I won’t hedge my deltas, but I’m betting on the divergence/convergence on vol levels and bank some coin via the premium within wing spreads.
Yes I know, anything within a months time frame is a pure gamma play, but I use several metrics for my bias for vol time frames and trade size, positions.
Cheers. 

