Two asset portfolio
Expected return:E(Rp) = xa Ra + xb Rb
= xa Ra +(1-xa)Rb
( as xa+xb =1)
Variance =σp2= xa2.σ a2+ xb2σ b2+ 2.ρab xa xbσaσb
A measure of how the returns on this asset move with respect to another asset
is given byc o v a r i a n c e, defined as
= COV ab =Σ pi. (Rai- Ra) (Rbi - Rb)
i= 1 to n
The Correlation coefficient, is expressed as
ρ = COV ab / σaσb
and can take a value from –1 to +1.
where Rai, Rbi are pairs of returns on shares `a’ and `b’ each assumed to be an
independent normally distributed random variable.
Ra is the expected or mean return on share `a’
Rb is the expected or mean return on share `b’
xa ,x b are the relative proportions of the total investment by value
in share a and b respectively and add to 1.
The relationship between the portfolio risk and the individual risks of the two shares
shows that it is possible to minimise portfolio variance through choosing shares which
have low correlation or ideally negative correlation asρ can take a value from +1 to-1
Note that ifρab = -1 ;
Thenσp=| (xa.σ a- xbσ b)|
And that further if
xa.σa = x bσb;
thenσp = 0.
The covariance of most stocks is positive and below 1; the return of a few stocks
have negative covariance. We can see from above that the risk reduction effect
ρab takes a value below 1, and is greatly enhanced when it has a negative value.