|
This paper finds that the return of the Taiwan''s stock
market are not normally distributed during the July 1991-June
1997 time period. In the presence of skewness, the portfoilo
selection is a trade-off competing and conflicting objectives,
such as maximizing both its expected returns and skewness, and
minimizing its risk for risk-aversion investors. Since it is
unlikely that a portfolio can solve the multiple-objectives
problem simultaneously, a portfolio selection must depend on the
investor''s preference among objectives. Polynominal goal
programming is utilized to determine the optimal portfolio
consisting of the choices of Taiwam''s 8 category stock indexes.
The features of applying polynomial goal programming in
portfolio selection are 1. the existence of an optimal solution,
2. the flexibility of the incorporation of investor preference
for skewness, and 3. the relative simplicity of computational
requirements. The empirical findings suggest that the
incorporation of skewness into an investor''s portfolio decision
causes a major chang in the construction of the optimal
portfolio. The evidence also indicate that investors trade
expected return of the portfolio for skewness. As was
expected, This study finds that efficient frontiers of no-short-
sales portfolio lie within the frontiers of short-sales
portfolio. Thus, with short sales allowed, investors will be
better off, as they are able to achive a higher level of
satisfation.
|