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

• 12/08/2020 at 7:08 PM #152990
  babula

  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

  

  babula

  Participant

  Topics: 5

  Replies: 0

  Been thanked: 0 times
  12/09/2020 at 2:23 PM #153090
  Nicolas

  You are talking about Covariance and Variance of a data serie:

  COVARIANCE

  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

  1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

  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

   

   

  

  Nicolas

  Keymaster

  Topics: 81

  Replies: 20863

  Been thanked: 4773 times
  12/09/2020 at 2:33 PM #153092
  Nicolas

  Variance indicator formula:

  VARIANCE

  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

  1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

  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

   

  

  Nicolas

  Keymaster

  Topics: 81

  Replies: 20863

  Been thanked: 4773 times
  08/19/2023 at 3:12 PM #219339
  LucasBest

  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

  1 2 3 4 5 6 7 8

  "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

  

  LucasBest

  Participant

  Topics: 16

  Replies: 208

  Been thanked: 72 times
  08/19/2023 at 3:18 PM #219340
  LucasBest

  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

  

  LucasBest

  Participant

  Topics: 16

  Replies: 208

  Been thanked: 72 times
  08/19/2023 at 3:21 PM #219342
  LucasBest

  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 ?
  

  

  LucasBest

  Participant

  Topics: 16

  Replies: 208

  Been thanked: 72 times
  08/19/2023 at 4:22 PM #219344
  JS

  Hi,

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

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

  1 user thanked author for this post.

  LucasBest

  

  JS

  Participant

  Topics: 26

  Replies: 1716

  Been thanked: 779 times
• Author
  Posts