Modeling Non-normality Using Multivariate t : Implications for Asset Pricing

@inproceedings{Kan2003ModelingNU,
  title={Modeling Non-normality Using Multivariate t : Implications for Asset Pricing},
  author={Raymond Kan and Guofu Zhou},
  year={2003}
}
In this paper, we propose to replace the widely used and firmly rejected multivariate normality assumption for the distribution of asset returns with a suitable multivariate t-distribution. Much of the asset pricing literature tries to explain expected returns on assets. In empirical studies, expected returns are often approximated by sample averages of realized returns. However, after replacing the normality assumption with a reasonable t-distribution, the most efficient estimator of expected… CONTINUE READING

References

Publications referenced by this paper.
Showing 1-10 of 28 references

Tests of mutlifactor pricing models, volatility bounds and portfolio performance, NBER working paper 9441

Ferson, E Wayne
2003

Economic implications of bull and bear regimes in UK stock returns, Working paper, University of Virginia and University of California, San Diego

Guidolin, Massimo, Allan Timmermann
2002

A general class of multivariate skew-elliptical

Branco, Márcia, Dipak K. Dey
2001

Mutual fund performance and seemingly unrelated assets ?

Ľuboš Pástor, Robert F. Stambaugh
2001

Some generalized tests of mean-variance efficiency and multifactor model performance, Working Paper, University of Pennsylvania

Geczy, C Christopher
2001

Comparing asset pricing models: an investment perspective

Pástor, L̆uboš, F Robert
Stambaugh, • 2000

Improved methods for tests of long-run abnormal stock returns

Lyon, D John, Brad M. Barber, Chih-Ling Tsai
Journal of Finance • 1999

Similar Papers

Loading similar papers…