st: mvrs ancillary files

[Kit Baum <baum@bc.edu>]

sacrificial piglet

Enzo said

in the SJ 7-1 prof Royston presented a new
multivariable regression spline approach (-mvrs-)
that looks, as usual, a very powerful new tool for Stata Users.
Unfortunately using the ancillary files the
examples described in the paper do not run
probably because the do files use some macro not
available on the web: -meanzero- , -gdelta-, -gph2wmf8- and -logopen- .



In running the examples from the three ado-files, I find that some of  
the examples for splinegen fail to work:

.  splinegen x, df(3) name(Z)
option name() not allowed
r(198);

. splinegen x, df(3) degree(1) orthog q(Q)
option q() not allowed
r(198);

Now it is true that neither name() nor q() appear in the definition  
of the splinegen, so these may just be out-of-date examples.


For the ancillary file brca_curves, make the following changes:

1) remove the Winjunk \a\c217\mvrs\ from the use statement

2) replace the 'meanzero f*, replace' line with

center f*

(if you do not have Ben Jann's -center-, ssc install center)

3) change the names of the fx* variables to c_fx* in the graphical  
statements

4) remove the lines gph2wmf8... to end; they are not needed. If you  
want the graph as a Windows metafile for some reason, you could use - 
graph export-.

The ancillary file will then run successfully.  Similar changes to  
the other do-files should render them usable. For instance, remove  
calls to 'logopen'; if you want a log, use the -log- command.

I have no idea what -gdelta- does. You'll have to ask Patrick (who I  
believe is not on Statalist).

30 lashes with a wet noodle for some associate editor of TSJ! You can  
be assured I was not responsible for checking out this submission.

Kit

Kit Baum, Boston College Economics and DIW Berlin
http://ideas.repec.org/e/pba1.html
An Introduction to Modern Econometrics Using Stata:
http://www.stata-press.com/books/imeus.html


*
*   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/