276 TIME-VARYING RETURN CORRELATIONS AND PORTFOLIOSare made, one for the total sample, one for the low-risk portfolios and one
for the high-risk portfolios.
With the complete sample of all portfolios, the DCC model portfolios has
significantly higherex postreturns than those of the rolling model. With
the low-risk portfolios the DCC model do not have a statistically significant
difference in returns as compared to the portfolios created using the rolling
model. On the other hand, the DCC model is clearly superior to the rolling
model for high-risk portfolios. In this case, the DCC model has returns
statistically significant from that of the rolling model for all the three periods
for which the comparison is made.
14.4 CONCLUSION
Even though the mean-variance optimization models have been around for
more than 50 years, practical uses of these models have been limited for
two reasons. Initially the model was not widely used due to the lack of
widespread availability of the computational power required for both the
estimation of variances and correlations, as well as the running of the opti-
mization model itself. With the advent of faster computers, this problem
has been considerably reduced in the past 20 years. The second and more
serious limitation of the model is the way the inputs into the model are esti-
mated. Until recently, computationally efficient multivariate models were
not available for estimating the co-variances between asset returns. With
the introduction of various multivariate GARCH models, this problem is
somewhat mitigated. This chapter has used one such model for estimat-
ing the co-variances to see whether portfolios created using these inputs
exhibit superior performance over those created with traditional estimates
of co-variances. The results indicate that the use of time-varying variances
and co-variances enhances theex postperformance of the efficient set of
portfolios.
REFERENCES
Best, M.J. and Grauer, R.R. (1991) “On the Sensitivity of Mean–Variance-Efficient Portfo-
lios to Changes in Asset Means: Some Analytical and Computational Results”,Review
of Financial Studies, 4(2): 315–42.
Bollerslev, T. (1990) “Modeling the Coherence in Short Run Nominal Exchange Rates:
A Multivariate Generalized ARCH Model”,Review of Economics and Statistics, 72(3):
498–505.
Chopra, V.K. and Ziemba, W.T. (1993) “The Effect of Errors in Means, Variances,
and Covariances on Optimal Portfolio Choice”, Journal of Portfolio Management,
19(2): 6–11.
Engle, R. (2002) “Dynamic Conditional Correlation: A Simple Class of Multivariate
GARCH Models”,Journal of Business and Economic Statistics, 20(3): 339–50.