st: Re: standardized coefficients from a probit model for path analysis using -listcoef-

[Hilary Taylor <Hilary.Taylor@bristol.ac.uk>]

Thanks for your kind reply Martin.  However my data has a binary dependent 
variable and ALL of the others (direct and indirect variables) are 
continuous, so this is not suitable for -ldecomp-.  Please correct me if 
I'm wrong.

If I use the syntax:

-ldecomp binaryvar, direct(contvar1) indirect(contvar2) or-

I receive the 'matsize too small' error

I'd be very grateful for any further help you, or anyone else on the list 
could offer.

Many thanks

Hilary

--On 13 September 2012 16:30 +0100 Hilary Taylor <ht4439@bristol.ac.uk> 
wrote:

> Hi, I would be very grateful if you would be able to help me with the
> following question:
>
> I am constructing a path analysis that has a binary outcome variable (all
> the other variables are continuous).  For the models with continuous
> outcome I am using multiple regression with the option -beta- to acquire
> fully standardized coefficients.  For the model with the binary outcome I
> am using -probit- and then following this up with -listcoef- (STB-57:
> sg152)and am reading from the 'bStdXY' column - the fully standardized
> coefficients.
>
> Could you tell me if my result from 'bStdXY' is the best equivalent for
> me to use in a model along with results from -beta-?
>
> Many thanks



----------------------
Hilary Taylor
Research Assistant
Academic Rheumatology
Musculoskeletal Research Unit
School of Clinical Sciences
University of Bristol
Avon Orthopaedic Centre
Southmead Hospital
Westbury-on-Trym,
Bristol. BS10 5NB

Tel: 0117 323 6187
*
*   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/