1. This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn More.

Chapter 20, Part I. Intro to Breakouts and Fakeouts Page 3

Discussion in 'Complete Trading Education- Forex Military School' started by Sive Morten, Dec 21, 2013.

Thread Status:
Not open for further replies.
  1. Sive Morten

    Sive Morten Special Consultant to the FPA

    Joined:
    Aug 28, 2009
    Messages:
    9,936
    Likes Received:
    10,634
    Commander in Pips: Ok, it’s better to show you formulas and examples. Let’s calculate volatility manually. In general volatility is a standard deviation of percent price changing from average price changing but not just a price from its average:

    Volatility = (Variance)^0.5,

    Variance = ∑(Xt –Xav)^2/(t-1)

    Where Xt – percent price change from previous value;

    Xav – average price change;

    t - Number of values.


    So i looks awful, but in reality this is rather simple.

    1. We need close prices, say, for the recent 10 trading sessions. You can use any number, but better to not use less than 5:


    [​IMG]

    2. Now we need to estimate percent changes of price. For that purpose we just need to divide followed number by previous. For instance 1.4380/1.4267 = 1.0079204 and so on:


    [​IMG]
    3. I prefer to you natural logarithm for calculation of percent deviations – here we just calculate natural logarithm from the numbers that we’ve estimated in #2. For instance, 100xLN(1.00790204)=0.788917416:


    [​IMG]

    4. Now we need calculate sum and mean values of logarithmic percent deviations, that’s simple:


    [​IMG]

    5. Now we calculate the numerator of variance formula – this is last column in our sheet. For example, the first row results as (0.788917416-0.1631231)^2 = 0.391618526:


    [​IMG]

    6. Calculate its sum…


    [​IMG]
    7. And in fact, our volatility is almost calculated, all that we need to do is to divide this number on (t-1) = 9, since we have 10 numbers and extract a square root, as on formula above:

    √(3.30203681/9) ~ 0.61%

    That’s daily volatility, since we’ve calculated it based on daily deviations. But how to express it in terms of weekly volatility or yearly volatility? Very simple, just multiply it on square root of days in week and/or year. So, weekly volatility will be: 0.61*√5 = 1.36%, while yearly = 0.61*√250 = 9.64%. The most common in application is yearly volatility, because it used in calculation of options value.

    Pipruit: Hm, you’re right, it does not look too scary, especially with Excel software. But, Sir, and why we use number of 5 for week and 250 for year?​

    Commander in Pips: Because, the common practice is to use number of business days in week so as in year. Still, some traders use 7days for the week and 360 or 365 for the year.

    Pipruit: And how to calculate yearly volatility, if you use weekly closes instead of daily?​

    Commander in Pips: Very simple. How many weeks in year?

    Pipruit: Approximately 52 weeks.
    Commander in Pips: So, you need to multiply weekly volatility on square root of 52 to get yearly one.

    Pipruit: Cool. And is there some software that could calculate this volatility automatically.​
     
    #1 Sive Morten, Dec 21, 2013
    Lasted edited by : Apr 23, 2016
Thread Status:
Not open for further replies.

Share This Page