Re: Why kullback-leibler distance?
- From: hrubin@xxxxxxxxxxxxxxxxxxxx (Herman Rubin)
- Date: 10 Apr 2006 10:21:14 -0400
In article <1144652191.738003.173780@xxxxxxxxxxxxxxxxxxxxxxxxxxxx>,
kalyan <mkalyan79@xxxxxxxxx> wrote:
Hi,
I am working with calculating the kullback-leibler distance between two
probability distribution. When this distance measure is not symmetric,
I was wondering why is the distance measure almost a standard to
calculate the distances of PDF ( probability distribution functions).
Or are there any distance measures which overcome this problem of
symmetry. I am calculating distances between multivariate gaussian
distributions.
Which distance measure is used depends on the purpose,
and really should be supplied by the one with the
problem. Kullback-Leibler may be two sensitive, even
when it is symmetric.
IMO, a good distance measure should be invariant under
change of base measure. K-L has this property, as
does L_1 and Hellinger. L_2 does not.
--
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, Department of Statistics, Purdue University
hrubin@xxxxxxxxxxxxxxx Phone: (765)494-6054 FAX: (765)494-0558
.
- References:
- Why kullback-leibler distance?
- From: kalyan
- Why kullback-leibler distance?
- Prev by Date: Re: [Q] Can a good random bit sequence generator be a good random number sequence generator?
- Next by Date: Re: Why kullback-leibler distance?
- Previous by thread: Re: Why kullback-leibler distance?
- Next by thread: logic for converting data obtained from input
- Index(es):
Relevant Pages
|
|
