Classical Portfolio Construction 237
FIGURE 7.4 The upper-right quadrant of the Cartesian plane
variance of returns in regards to defining potential risk. While this may very
well be true, we will not address this concept any further in this chapter so
as to discuss E–V theory in itsclassicsense. However, Markowitz himself
clearly stated that a portfolio derived from E–V theory is optimal only if the
utility, the “satisfaction,” of the investor is a function of expected return and
variance in expected return only. Markowitz indicated that investor utility
may very well encompass moments of the distribution higher than the first
two (which are what E–V theory addresses), such as skewness and kurtosis
of expected returns.
Potential risk is still a far broader and more nebulous thing than what
we have tried to define it as. Whether potential risk is simply variance on
a contrived sample, or is represented on a multidimensional hypercube,
or incorporates further moments of the distribution, we try to define po-
tential risk to account for our inability to really put our finger on it. That
said, we will go forward defining potential risk as the variance in expected
returns. However, we must not delude ourselves into thinking that risk
is simply defined as such. Risk is far broader, and its definition far more
elusive. There will be more on this in Chapter 12.
So the first step that an investor wishing to employ E–V theory must
make is to quantify his or her beliefs regarding the expected returns and
variance in returns of the securities under consideration for a certain time
horizon (holding period) specified by the investor. Theseparameterscan
be arrived at empirically. That is, the investor can examine the past history