Re: st: mm_root arguments limitation issue

[Austin Nichols <austinnichols@gmail.com>]

Ben Jann is the author of mm_root, distributed in -moremata- on SSC.
Note that the FAQ advises you to give the source of user-written
software.

Ben wrote me on another occasion:
"
the limit of 10 passthrough arguments comes from mm_callf() which is
used in mm_root(). But this is not really a limitation. In fact, one
single passthrough argument would suffice, since you can put
everything into one object and then have the called function take the
object apart. In you case, you could type something such as

function y(x, v)
{
   a=v[2]; b=v[3]; c=v[4]; d=v[5]; e=v[6]; f=v[7]
   g=v[8]; h=v[9]; i=v[10]; j=v[11]; k=v[12]
   return((-1*(a-b)*(1+x)^5) + ((c-(b*x))*(1+x)^4) +
       ((d-(e*x))*(1+x)^3) + ((f-(g*x))*(1+x)^2) +
       ((h-(i*x))*(1+x)) + ((j-(k*x))) + ((j-(k*x))/x))
}
for (i=1; i<=rows(v); i++) {
   r = mm_root(x=., &y(), 0, 1, 0, 1000, v[i,.])
   v[i,1] = x
}

In more complicated situations, v could be a pointer vector or a
structure containing different types of elements.

[One may face] a problem with solutions outside the specified search
interval. In such a case mm_root() will issue an appropriate return
code that can be used to trigger another search with a different
interval. Say you search over interval [0,100] and mm_root() returns
code 3 then you could have your program start a new search, say,
between 100 and 200. And so on. For example, code something like

lo = 0
up = 100
while (1) {
   r = mm_root(x=., &y(), lo, up, ...)
   if (r=2) {
       lo0 = lo
       lo = lo - (up-lo)
       up = lo0
   }
   else if (r=3) {
       up0 = up
       up = up + (up-lo)
       lo = up0
   }
   else {
       break
   }
}

"
On Thu, Sep 1, 2011 at 1:38 AM, dai_yunhao <dai_yunhao@smail.hust.edu.cn> wrote:
> Hi everybody
>
> I'm a new user of stata, and  facing the problem about the mm_root arguments
> limitation issue.
> I try to find the source code of  mm_root and copy the code to a new file,
> for the issue, I change the first line of the source code
>  local opts "o1, o2, o3, o4, o5, o6, o7, o8, o9, o10 "
> to
>  local opts "o1, o2, o3, o4, o5, o6, o7, o8, o9,
> o10,o11,o12,o13,o14,o15,o16,o17,o18,o19,o20,o21,o22,o23,o24,o25,o26,o27,o28,o29,o30"
>
> My question is :
> 1 Do I correctly change the code?
> 2 Which format should I save the new file to ? Should I click the menu
> file-->save as-->filename.ado?
>
> Thank you!
>
> Dai Yunhao
>
>

*
*   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/