Explain kurtosis, Advanced Statistics

Kurtosis: The extent to which the peak of the unimodal probability distribution or the frequency distribution departs from its shape of the normal distribution, by either being more pointed (like leptokurtic)or flatter ( like platykurtic). Commonly measured for a probability distribution as

745_kurtosis.png 

where 4 is the fourth central moment of distribution, and 2 is its variance.

(consequent functions of sample moments are used for frequency distributions.)

For the normal distribution this index takes the value three and often index is redefined as the value above minus three so that the normal distribution would contain the value zero.

(Other distributions with the zero kurtosis are known as mesokurtic.) For the distribution which is leptokurtic the index is positive and for the platykurtic curves it is negative. It is shown in the figure 

26_kurtosis1.png

Posted Date: 7/30/2012 1:36:00 AM | Location : United States







Related Discussions:- Explain kurtosis, Assignment Help, Ask Question on Explain kurtosis, Get Answer, Expert's Help, Explain kurtosis Discussions

Write discussion on Explain kurtosis
Your posts are moderated
Related Questions
The risk of being able to recognize the respondent's confidential information in the data set. Number of approaches has been proposed to measure the disclosure risk some of which c

wat iz z difference b/n logistic regression and multiple regression analysis /

Human capital model : The model for evaluating the economic implication of the disease in terms of the economic loss of a person succumbing to morbidity or the mortality at some pa

Q1: The growth in bad debt expense for Aptara Pvt. Ltd. Company over the last 20 years is as follows. 1997 0.11 1998 0.09 1999 0.08 2000 0.08 2001 0.1 2002 0.11 2003 0.12 2004 0.1

Hazard plotting  is based on the hazard function of a distribution, this procedure gives estimates of distribution parameters, the proportion of units failing by the given time per

Locally weighted regression  is the method of regression analysis in which the polynomials of degree one (linear) or two (quadratic) are used to approximate regression function in

Obuchowski and Rockette method  is an alternative to the Dorfman-Berbaum-Metz technique for analyzing multiple reader receiver operating curve data. Instead of the modelling the ja

VIF is the abbreviation of variance inflation factor which is a measure of the amount of multicollinearity that exists in a set of multiple regression variables. *The VIF value

difference between histogram and historigram

Multitrait multi method model (MTMM) is the form of confirmatory factor analysis model in which the different techniques of measurement are used to measure each of the latent vari