True. However if you are modelling / simulating, heavy tails may be called for in light of observations or theory as opposed to conventional hypothesis testing.
Herman Rubin wrote: > In article <[EMAIL PROTECTED]>, > Robert Ehrlich <[EMAIL PROTECTED]> wrote: > >Apparently not; according to Evans, Hastings, and Peacock " the LaPlace provides > >a heavier tailed alternative to the Gaussian." (Statistical Distributions, > >Wiley) > > I cannot see anyone making such a statement. There is no > particular reason why data should follow any particular > type of distribution in general, and this includes the > normal. In particular cases, one might be able to justify > a distribution such as the binomial or even the Laplace > distribution on theoretical grounds, but I do not know of > a "natural" model which will yield the normal, except as an > approximation. > > >Chia C Chong wrote: > > >> Hello... > > >> I wonder, are there any mathematical relationship between the Gaussian & > >> Laplacian PDF? How about the statistical explaination of these twoPDFs? I am > >> very interested to know more on these 2 PDFs. I would be grateful if someone > >> have ever come across any articles that discuss these would let me know the > >> further details. > > >> Thanks. > >> CCC > > -- > This address is for information only. I do not claim that these views > are those of the Statistics Department or of Purdue University. > Herman Rubin, Dept. of Statistics, Purdue Univ., West Lafayette IN47907-1399 > [EMAIL PROTECTED] Phone: (765)494-6054 FAX: (765)494-0558 . . ================================================================= Instructions for joining and leaving this list, remarks about the problem of INAPPROPRIATE MESSAGES, and archives are available at: . http://jse.stat.ncsu.edu/ . =================================================================
