Re: eliminate noise from a regression model
- From: dave@xxxxxxxxxxx
- Date: 10 Feb 2006 14:01:48 -0800
DESI,
Perhaps you have a case where the parameters "differ" for different
subsections of the data ... This can and often does happen with time
series data . In cross-sectional data this can also be present when an
unspecified variable has changed . This is one reason why statisticians
randomize.
Lots of noise can be a resultant of the need for a variance-stabilizing
transformation.
Lots of noise can simply be a manifestation that the candidate X's are
just not important or that the right X's haven't been included .
HTH
Dave Reilly
Automatic Forecasting systems
http://www.autobox.com
.
- References:
- eliminate noise from a regression model
- From: desigun
- eliminate noise from a regression model
- Prev by Date: eliminate noise from a regression model
- Next by Date: Re: Mds
- Previous by thread: eliminate noise from a regression model
- Next by thread: Re: eliminate noise from a regression model
- Index(es):
Relevant Pages
|
|
