how to calculate the standard deviation of a portfolio when no correlation is given?

CFA Level III
#1

76% in portfolio A with SD of 5.8% and 24% in portfolio B with SD of 7.8%.

I calculated as the square root of (0.76 square * 0.058 square + 0.24 square * 0.078 square)

Schweser simple takes the weighted average of the two: 0.76*0.058 + 0.24*0.078=6.28%

I remeber CFA uses my approach…

#2

, gave a wrong answer here … corrected below.

#3

Your approach assumes correlation coefficent is zero. They assume correlation coefficient=1.

Was anything else shared? Im curious why the rational of one over the other.

Keep in mind sometimes a problem might give other information that one can calulate the cc. For example covarience, betas and standard deviation of the market, etc.

#4

Schweser 2015 pracice exam volume 1 exam 2 morning session question 6.

I don’t think I missed anything…

#5

if it was a corner portfolio problem - schweser was correct with w1*sd1 + w2*sd2. (assuming a correlation coefficient of 1 the way gad4 mentions in his post).

#6

I see.

So for corner portfolio, we always assume a correlation coefficient of 1?

Thanks!

#7

It’s a conservative approach; we really think that it’s less than +1.0, but assuming +1.0 gives us the maximum possible value for the variance.