Table 4. It is apparent from the results that the extension of standard CAPM by incorporating
conditional coskewness has improved the results. The intercept term is significantly different from zero
in sub-periods 1993-1995, 1993-1998 and 1999-2004. The premium for beta risk is also positive and
significant for the period 1993-1995 and inconclusive and insignificant otherwise. The price of
conditional coskewness risk is significantly different from zero in sub periods 1993-1995, 1996-1998
and 1993-1998 and the overall sample period. 1993-2004. The risk premium for conditional cokurtosis
risk when it is taken as an additional explanatory variable with covariance risk is positive and
significant in sub-periods 1993-1995 and 1993-1998 (with monthly data). It is inconclusive and
insignificant in other sub-periods and overall period. The intercept term remain significantly different
from zero for most of the sub-periods and overall sample period except for the sub-period 1993-1995
and 1993-1998. The results remain the same for four moment CAPM model. The beta risk is positively
and significantly compensated only for the period 1993-1995. These results indicate that covariance
and cokurtosis risk have limited compensation only for few periods, but investors get reward for
conditional coskewness risk in the Karachi stock Market. The evidence on conditional higher moment
asset pricing model by alternative methodology where the risk parameters βi, γi and κi are estimated by
cubic model in the first stage by GMM procedure and the findings reported in appendix Table A2 are
similar to those reported in Table 4.