But is it a different ( regression or panel regression) case when you use
explanatory variables that have a different natural scale?
I.e. explanatroy variable a ranges from (0;2) and variable b from
(1'000'000; 8'000'000). Would this in some way distort the estimation or
would coefficients just be scaled accordingly ? Or should you scale /
standardize variables in some way (e.g. dividing variable b by 1Mil.) ?
- Tom
Von: [email protected]
[mailto:[email protected]] Im Auftrag von Nick Cox
Gesendet: Mittwoch, 10. Januar 2007 19:18
An: [email protected]
Betreff: st: RE: scaling
Your regression coefficients have units that
are units of response / units of predictor. Thus
if you don't like them change one unit or the other
or both until the numbers are congenial.
Nick
[email protected]
Renuka Metcalfe
> I am using a panel data. I am looking at the impact of
> "x" on employment growth, runing an OLS. I ran
> something like
>
> .regress empgrowth2004 x1, x2, x3
>
> Some of my coefficients are very large. For example,
> 1307.
>
>
> I would be grateful, if anyone would let me know, how
> I could scale these coefficients so that they are a
> lot smaller.
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
