Think of what you are asking Mata to do.
st_store(1, ...)
means: "Write something to observation 1." If you type
st_store(1, "x", "rc", x)
(or -st_store(1, "x", 1, x)- if rc is the first variable) and variable
"rc" is zero for the first observation, this means: "Write x to no
observation". "no observation" has zero rows, but x is a scalar in
your example and therefore has 1 row. Hence the conformability error.
You could code, e.g.
rc = (mm_root(x = .,&myfunc(),0,1,0,1000,a)==0)
if (rc) st_store(1, "x", x)
ben
On Wed, Feb 11, 2009 at 4:25 PM, Bob Hammond <[email protected]> wrote:
> All,
>
> I want to store a Mata scalar in my Stata data set using st_store(), but
> only in certain cases, as determined by selectvar. My reading of the
> st_data() help file (where st_store() directs you to concerning selectvar)
> says that the following should work:
>
> clear
> cap mata: mata drop myfunc()
> set obs 1
> gen double rc = .
> gen double x = .
> mata
> function myfunc(x, a) return(x^2 - a)
> a = 2
> rc = (mm_root(x = .,&myfunc(),0,1,0,1000,a)==0)
> st_store(1,"rc",rc)
> st_store(1,"x",1,x)
> end
> list
>
> My Stata variable "rc" is equal to one if the mm_root() function has found a
> valid solution. I only want st_store() to write the solution to the data
> set when it is valid.
>
> Instead, the program correctly writes the solution to the data set when it
> is valid (try by changing a to "a = 2/3" above) but generates a "3200
> conformability error" when it is invalid (try the program as written above,
> which generates no valid solution because it looks between 0 and 1 for the
> square root of 2).
> Maybe this is a simple error in my reading of how to use selectvar but I
> have tried it by placing "rc" and, alternatively, by placing 1 (the variable
> number of "rc") in the third argument of st_store(). Both ways give the
> same conformability error.
>
> Any thoughts would be appreciated. Thanks in advance,
>
> Bob
> --
> ------------------------------------------------------------------------
> Bob Hammond
> Department of Economics
> North Carolina State University
> Office: (919) 513-2871
> Fax: (919) 515-7873
> http://www4.ncsu.edu/~rghammon/
> *
> * 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/
>
*
* 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/
