Re: st: relative importance of independent variables

[Richard Williams <richardwilliams.ndu@gmail.com>]

At 07:26 AM 6/14/2011, Maarten Buis wrote:
> On Tue, Jun 14, 2011 at 2:04 PM, Stefan Nijssen wrote:
> > I am running an OLS, see below, in which I relate 3 variables 
> (currentratio, margin and da) to OAS, all continuous. I control for 
> industry and country (twice a triple dummy, added to the equation). 
> Seems to perform reasonably well to me. However, I have one last 
> issue to overcome. I would like to determine the relative 
> importance to the model, and especially the hierarchy of this 
> importance, of the three independent variables currentratio, margin 
> and da. Probably this is task must be doable but I cant seem to 
> find a proper way through any academic articles or google.
>
> This is a problem of standardization; you must make sure that these
> three variables have the same unit. A common strategy is to divide
> each variable with its standard deviation, so now all effects are the
> effects of one standard deviation increase. See e.g. the -beta- option
> in -regress-. Whether that makes sense is a substantive question.
>
> Hope this helps,
> Maarten

The -beta- option will standardize the dependent variable as well. 
X-Standardization alone may be sufficient. The -listcoef- command, 
which is part of the Long and Freese -spost9- package (use -findit-) 
has handy options for various kinds of standardization.


-------------------------------------------
Richard Williams, Notre Dame Dept of Sociology
OFFICE: (574)631-6668, (574)631-6463
HOME:   (574)289-5227
EMAIL:  Richard.A.Williams.5@ND.Edu
WWW:    http://www.nd.edu/~rwilliam

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/