It is possible to code the beta(cov(x,,y)/var(y))? variance and covariance

babula · 2020-12-08T18:08:02+00:00

I can´t how to create the beta indicator cause within the indicator i don´t know how to get the var of the marekt and also how to get the cov (stock,market). Thx

Viewing 7 posts - 1 through 7 (of 7 total)

Author

Posts
12/08/2020 at 7:08 PM #152990 quote

babula
Participant

New

I can´t how to create the beta indicator cause within the indicator i don´t know how to get the var of the marekt and also how to get the cov (stock,market). Thx

12/09/2020 at 2:23 PM #153090 quote
Nicolas
Keymaster

Legend
You are talking about Covariance and Variance of a data serie:
```
length = 10
srca = open
srcb = close
a = ((srca-srca[1])/srca[1]*100)
b = ((srcb-srcb[1])/srcb[1]*100)
aavg = average[length](a)
bavg = average[length](b)
sum = 0
for i = length downto 0 do 
 sum=sum+((a[i]-aavg)*(b[i]-bavg))
next 

covariance = sum/(length-1)

return covariance
```
12/09/2020 at 2:33 PM #153092 quote
Nicolas
Keymaster

Legend
Variance indicator formula:
```
t   = 10 //Lookback
k   = 2 //Single VRT Sampling Period
src = customclose

x = log(src/src[1])
xk = log(src/src[k])
mean = average[t](x)
sum = 0
for i = 0 to t-1-k do 
sum=sum+(square(xk[i]-mean*k))
next
m = k * (t - k + 1) * (1 - (k/t))
variance = sum/m

return variance
```
08/19/2023 at 3:12 PM #219339 quote
LucasBest
Participant

Senior
How is it possible to translate this instruction from Pine script, as there is only one period (length) and not 2 like in the above script for Variance ?
```
"ta.variance(source,length)"

Also, as standard deviation equal the square root of the variance.
Is it possible to calculate more quickly and simply the Variance with std squared ?

Variance(src,length)=STD[length](src)*STD[length](src)

Thanks
```
08/19/2023 at 3:18 PM #219340 quote

LucasBest
Participant

Senior

standard deviation is simply obtained by calculating the square root of the variance, is it possible to calculate more quickly and simply the Variance with std squared Thanks

08/19/2023 at 3:21 PM #219342 quote

LucasBest
Participant

Senior

Sorry, i don”t understand why my reply appears like they did… As STD is simply obtained by calculating the square root of the variance, is it possible to calculate more quickly and simply the Variance with std squared ?

08/19/2023 at 4:22 PM #219344 quote

JS
Participant

Master

Hi,

That’s right, the variance can be found by squaring the standard deviation…

Variance= Square(Std[x](Close))

LucasBest thanked this post
Author

Posts