Why Do We Care About Volatility? part 3: Calculating implied volatility
All volatility figures are quoted on an annualized basis unless stated otherwise. Since we did not state otherwise in last weeks example, the marketplace thinks that "most likely" the underlying stock will not be below 80 or above 120 at the end of one year. "Most likely" is defined as in 68% of all occurrences.
Our next move is to break down the one standard deviation move so that it fits any time period -- that way you can apply it to any expirations you're interested in trading. The formula in the simplest form is the following:
OSDM = Price x Implied Volatility (ann.) x (sqrt) Calendar Days to Exp.
____________________________________________________________
(sqrt) 365
(A more accurate formula is to use trading days to expiration instead of calendar days, and then divide the entire equation by (sqrt) 252, which is the total number of trading days in a year.)
Now let's assume we are dealing with a 30 trading day option contract. The one standard deviation move then becomes:
OSDM = 100 x .20 x (sqrt) 30
________________
(sqrt)365
OSDM = + 5.73
This means over a 30 calendar day period the underlying is expected to finish between 94.27 and 105.73. If we perform the calculations for a 60 and 90 calendar day period and then graphed the results the graph would look like the following:

This shows us that the longer the time period, the larger the potential for wider swings in the underlying stock. Keep in mind that even if you are looking at a 30, 60, or 90 day options the implied volatility will always be quoted as an annualized number.
Regards,
Brian (OG)
While implied volatility represents the consensus of the marketplace as to the future level of stock price volatility there is no guarantee that this forecast will be correct.
Edited by optionsguy at 10/07/08 at 03:20 PM


Comments
Follow commentshttp://community.tradeking.com/members/ posted February 08, 2006 (07:00PM)
Hi Brian,Great Blog!! Very useful information. Please continue educating us. I'm reading. Looking forward to much more.
Thank You,
Rich
optionsguy posted February 12, 2006 (07:00PM)
Brian, you are right about the difference between normal and lognormal - except the normal is RETURN SPACE! So a 20 volatility is RETURNS up and down of 20%. Now assuming the future value is 100, 68% between 100*(1.20) and 100/1.20 – this implies a one Standard Deviation (SD) move lognormally to be between 83.33 and 120, the two SD to be between 71.42 and 140. Understanding that the options pricing assumption is based on continuously compounded returns we could use e^.20 and e^-.20 to be more precise. The calculations then determine the one SD move to be between 81.87 and 122.14, and two SD move to be between 67.03 and 149.18.This now lets us understand 100 volatility:
If your future value is 100, a one SD at 100 volatility is between 50 (which is 1/2 of 100) and 200 (which is 2*100) which is exactly how the lognormal distribution works. With that be said, from an elementary perspective with low volatility and short time, price rather than return is fine.
optionsguy posted February 12, 2006 (07:00PM)
Hello All,I received a note from a mentor of mine in the field of options that will shed some advanced flavor to the above blog (below comment). He took the time to explain and calculate the difference between the normal and lognormal distributions. He sent this information to me via an email and I thought since he took the time that I would share his math with everyone. To the advanced option trading geeks I say – Hear, Hear!
Regards,
Options Guy
(regarding comment directly below)
http://community.tradeking.com/members/ posted March 21, 2007 (08:00PM)
Well, lost you again. When I take 100 multiply it by .20 and multiply that by 30 and then divide that bu 365, I get 1.6438. Please, how did you get 5.73? What am I missing?Thanks for your help.
Brodawg posted February 18, 2008 (05:01AM)
Brian, I'm with "Anonymous" on this one. How do you get 5.73 when you do your OSDM equation.
(100 x .20 x 30) / 365 = 1.6438
-Brodawg
optionsguy posted February 18, 2008 (05:48AM)
Hello Brodawg and Anonymous,Both of you guys are ignoring the (sqrt) before the 30 and the 365 days. "Sqrt" was intended to mean "square root" of 30 (5.4772255) and square root of 365 (19.104973). This is because options decay at the square root of time, so when you are dealing with days to expiration we take the square root of the number. If you look at the Black/Scholes option pricing model written out you will see a square root of T in the equation.
Regards,
Brian (Og)
You must Log In to post to this blog.
Not a member? Register Now to …