0
$\begingroup$

So my question goes like this, I have the returns of 3 different stocks AAPL, NKE and BBRY I make 4 portfolios out of them as follows:

my stock portfolios

and the question asks me to compute the correlation coefficient and covariance for each portfolio...

Now this seems a bit odd to me, since it already asked me to compute the covariance and correlation of each stock pair in an earlier question. Also, how would I go about doing it for the portfolio with 3 stocks?

$\endgroup$
  • 1
    $\begingroup$ Hint: The difference is the 1) portfolios are weighted unevenly between stocks, and 2) you can't just sum the variances to get the whole. You need to add covariances also.. $\endgroup$ – RegressForward Apr 5 '15 at 15:35
  • $\begingroup$ So am I suppose to find the weighted variance of the portfolio and use that as the covariance? i.e. $$ Cov[Portfolio] = \sum_{i=0}^n \sigma_i^2w_i^2 + \sum_{i=0}^n\sum_{k\ne i} w_iw_k\sigma_{i,k}\sigma_i\sigma_k $$ So like for the 3 stock portfolio I will be using $$ Cov[Portfolio] = \sum_{i=0}^3 \sigma_n^2w_n^2 + 2w_1w_2\sigma_{1,2}\sigma_1\sigma_2+ 2w_1w_3\sigma_{1,3}\sigma_1\sigma_3+ 2w_2w_3\sigma_{3,2}\sigma_3\sigma_2 $$ $\endgroup$ – asosnovsky Apr 5 '15 at 17:15
  • 1
    $\begingroup$ Looks great to me! $\endgroup$ – RegressForward Apr 6 '15 at 13:43
  • $\begingroup$ Now how would I get the correlation? $\endgroup$ – asosnovsky Apr 6 '15 at 13:59
  • $\begingroup$ The correlation can be found if you know the covariance. math.stackexchange.com/questions/186959/… $\endgroup$ – RegressForward Apr 6 '15 at 14:41

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.