kurtosis

(redirected from Leptokurtic distribution)
Also found in: Dictionary, Financial, Encyclopedia.
Related to Leptokurtic distribution: Mesokurtic, skewness, Excess kurtosis

kurtosis

 [ker-to´sis]
the degree of peakedness or flatness of a probability distribution, relative to the normal distribution with the same variance. See illustration.
Kurtosis. From Dorland's, 2000.
Miller-Keane Encyclopedia and Dictionary of Medicine, Nursing, and Allied Health, Seventh Edition. © 2003 by Saunders, an imprint of Elsevier, Inc. All rights reserved.

kur·to·sis

(kŭr-tō'sis),
The extent to which a unimodal distribution is peaked.
[G., an arching]
Farlex Partner Medical Dictionary © Farlex 2012
References in periodicals archive ?
The recent financial crisis has underscored the importance of active responses to extreme tail risks that are associated with a leptokurtic distribution of key monetary variables and stem from their volatility outbursts (Orlowski, 2010b).
If the estimated GED parameter is less than 2, leptokurtic distribution is detected.
This is the defining property of leptokurtic distributions and is the fundamental reason why the counter-intuitive approach adopted here seems to produce a low risk portfolio according to a required rate of return, with far fewer stocks than the mean-variance approach.
It would make sense that a stock in a sector that experiences boom and bust cycles would have underlying leptokurtic distributions of returns.
(2004) survey several variations of GARCH modeling and they conclude that leptokurtic distributions, especially Student-t, are more adequate than the Normal distribution to be used in ARCH and GARCH-family models.
With respect to kurtosis, all the markets demonstrated leptokurtic distributions. The histograms in Appendix B (see Figures B1-B7) illustrate the results along with the normal standardized distributions corresponding to the samples' means and variances.
In many empirical studies, the standard Student-t distribution has been found to be quite successful in modeling highly leptokurtic distributions (e.g., Blattberg and Gonedes, 1974; Bollerslev, 1987 and Kon, 1994).
This occurs because extremely leptokurtic distributions of effects decrease the frequency of mutations with effects in the class that is most damaging (i.e., s [congruent]1/[N.sub.e]).
Spatial contact models incorporate a variety of contact distributions, including the leptokurtic distributions that are so prevalent in the dispersal data.