Forums  > Pricing & Modelling  > Hourly to Daily Volatility Conversion  
     
Page 1 of 1
Display using:  

ps1212


Total Posts: 3
Joined: Sep 2017
 
Posted: 2022-07-29 20:32
Most security price volatility seems to be measured using daily data to create a daily volatility, then it can be converted to monthly or annual by multiplying by sqrt(12) and sqrt(252) respectively.

If my starting point is hourly data instead of daily data the same technique would convert S&P 500 futures hourly volatility to daily by multiplying by something in the range of sqrt(6.5) to sqrt(24) with something close to the low end of the range possibly making the most sense to me initially since there are 6.5 hours in the cash equity day and those are the most liquid and active hours for the futures.

When I do this however, multiplying the hourly volatility by anything makes the value much higher than the daily data - at least for the periods I've explored. If I do multiply by sqrt(6.5) then the daily volatility derived from the hourly volatility seems way too high.

Can someone guide me here on the correct way to convert hourly volatility to a daily volatility number?

ConditionalExpectations


Total Posts: 12
Joined: Jul 2021
 
Posted: 2022-07-30 03:40
Your formula is correct for writing hourly volatility in the same units as daily volatility. This is not the daily volatility of the security; it is simply the hourly volatility of your security, represented with different units.

There cannot exist a formula to convert hourly volatility to daily volatility. Imagine a security with tremendous hourly volatility that resets to the same price at the end of each day, and a security with similar hourly volatility that reaches a new high at the end of each day; any such formula would assign the same daily volatility to both securities, a contradiction.

When you say that this formula gives a quantity that seems way too high, could it be because you are comparing it to the properly calculated daily volatility (a completely different quantity)?

Silence Dogood No. 4

nikol


Total Posts: 1457
Joined: Jun 2005
 
Posted: 2022-07-30 08:08
A structure in transition probabilities of price evolution or saying differently presence of autoregressions will inevitably lead to deviation from pure sqrt.

Your finding indicates that autocorrelation coeff is negative.

https://en.m.wikipedia.org/wiki/Autoregressive_model

... What is a man
If his chief good and market of his time
Be but to sleep and feed? (c)

articlepost


Total Posts: 1
Joined: Aug 2022
 
Posted: 2022-08-18 03:25
Choice is a very interesting thing.
Angelia pursues a plain life, so she has the courage to abandon the high-end life pursued by thousands of girls.
We cannot subjectively judge whether the choices made by the two heroines are correct or not. What we can see is that they have bravely chosen the life they want.
but on whether what she pursues is what she likes, and what she gives up is what prevents her from realizing her dream.
VoIP virtual phone number
A very interesting website
Previous Thread :: Next Thread 
Page 1 of 1