How the fuck do I compute the volatility of a stock price?
Say I have a vector with a stock's price over n days. How do I best calculate it?
Right now I just calculate the average price over those n days, then for each day I do:
(Price - Average)^2
I sum all of those, then take the average of that, and then the square root of that.
What I then get I divide by the average price, and that's what I use as volatility.
I chose this method because I found it the most intuitive, however I have no actual clue as to how it should really be done. Can /biz/ please help me?
>>1274408
Bollinger bands with either simple moving average or exponential moving average.
daily expected volatility for the year for aapl is .05%
you should expect to see it trading between 80 and 120 next year this time
Alright boys I figured out how to do it, if anyone's interested I will post the method.
>>1274408
>>1274422
If you could mathematically compute "risk" and account for it, you would have solved the market and introduced a new paradigm in global trade.
>>1274681
Fuck off with your smug comment pointing out the obvious and thinking you are all smart for it, he asked for a volatility metric for his data.
Learn to read, you fucking retard.
>>1274686
No such thing as a volatility metric. You'd have to be able to quantify risk, which none of you chuckle fucks can get your heads around.
LOL, technical analysis people are bottom feeders.
>>1274689
You literally don't have any idea of what you are on about.
>No such thing as a volatility metric
- Yes; Variance, Standard Deviation.
>You'd have to be able to quantify risk
- Which is done in one of the most basic theories of finance, the risk premium.
>>1274682
>This is the standard deviation which will also work as a metric for volatility.
I found another way now, I do it with a log approach.
Basically we create new values x_i = ln(P_i/P_i-1). Then we calculate the mean of those, and THEN compute the standard deviation of those x_i.
After that, because I have the vector with prices on a daily basis, we multiply that by the square root of 365.
i mean, literally, you can find daily expected volatility by looking at the option's implied volatility
>trying to use mediocristan metrics in extremistan environments
havent you read black swan by NNT?
okay
CMG
current IV: 28.35%
28.35/19.1=1.48% expected daily volatility
that's about $6.59 per day it's expected to move in either direction
$571 and $319 is the expected range CMG should close in a year
All the work has already been done for you. Just check the beta and the peg ratio
>>1275066
I've read it, I was just trying to play around a bit with the Black Scholes equation in MATLAB.
However, how does our friend NNT compute option prices?
>>1274422
this is correct. standard deviation of the price is stock price volatility.
>>1275946
>that is return variance and standard deviation.
I found this method in the book "Option Volatility and Pricing" by Sheldon Natenberg. I should use the other way to calculate it?