But this is faster, as it reduces the
interpretive overhead of ploughing through
-egen- repetitively.
unab myvars : price-foreign
qui foreach v of local myvars {
local others : list myvars - v
tokenize `others'
gen max_`v' = `1'
local others : list others - 1
foreach w of local others {
replace max_`v' = max(max_`v', `w')
}
}
Nick
[email protected]
Nick Cox
> unab myvars : ab-AB
> foreach v of local myvars {
> local others : list myvars - v
> egen max_`v' = rowmax(`others')
> }
Steve Vaisey
> > I have been wrestling with a problem for a while, and I've finally
> > decided to bring it to you for what will probably be an
> embarassingly
> > quick solution. Here's the problem. Thanks to Nick, I have
> a program
> > (-myboolean-) that turns fuzzy sets into multiset fuzzy
> > intersections.
> > So, for example, with fuzzy sets A and B, running -myboolean A B-
> > yields four new variables: ab Ab aB AB (where lower = absence
> > [1-set],
> > upper = presence [set], and -- for example -- AB =
> min(A,B)). Enough
> > background I hope.
> >
> > What I'd like to do now is to create new variables that contain the
> > maximum value of all OTHER configurations for each configuration
> > produced by -myboolean-. Manually, this is easy to do: otherAB =
> > max(Ab,aB,ab). But with 32 or 64 configuration variables,
> > this is very
> > time consuming. Is there a way to automate this so that:
> > foreach var of varlist ab - AB {
> > gen other`var' = max(***all other vars in varlist but `var'***)
> > }
> >
> > One final note: in the crisp set case (where set membership
> is always
> > 0/1) other`var' would always = 1-`var'. But this is not true in the
> > fuzzy case. Hence the difficulty.
*
* 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/
