hi all,
we try to familiarize ourselves with the newly added Mata structures but we
got stuck. In the very end we would like program "hashing" in order to make
searching more efficient. The kind of problem we have can be described as
follows.
We have defined the mata structure "point":
struct point {
real scalar a
real scalar b
}
and the function "help" that should return a vector of points. The
function is (or should be) defined such, that if it receives the real vector
seq =4,5 the return vector of point should be (4,4),(5,5).
The Mata Code for the function "help" looks as follows:
struct point vector function help(real vector seq) {
real scalar length
length = length(seq)
struct point vector v
v = point(length)
real scalar i
struct point scalar mypoint
for (i=1; i<=length; i++) {
mypoint.a=seq[i]
mypoint.b=seq[i]
liststruct(mypoint)
v[i] = mypoint
liststruct(v)
}
return(v)
}
Now, let's run this function by parsing the row-vector 4,5 to it:
: seq = 4,5
: help(seq)
Here is a commented log of the output: The output of -liststruct(mypoint)- in
the first iteration of the loop was:
1 structure of 2 elements
1.1 1 x 1 real = 4
1.2 1 x 1 real = 4
1 1 x 2 structure of 2 elements
and the output of -liststruct(v)- in the first iteration of the loop was:
[1,1]:
1.1 1 x 1 real = 4
1.2 1 x 1 real = 4
1 [1,2]:
1.1 1 x 1 real = .
1.2 1 x 1 real = .
So far, this is the desired result. The problem arises in the second iteration
of the loop. The output of -liststruct(mypoint)- in the second iteration of
the loop was:
POINT2: (5,5)
1 structure of 2 elements
1.1 1 x 1 real = 5
1.2 1 x 1 real = 5
1 1 x 2 structure of 2 elements
which again is what we expected. But the output of -liststruct(v)- was
[1,1]:
1.1 1 x 1 real = 5
1.2 1 x 1 real = 5
1 [1,2]:
1.1 1 x 1 real = 5
1.2 1 x 1 real = 5
Although we had expected
[1,1]:
1.1 1 x 1 real = 4
1.2 1 x 1 real = 4
1 [1,2]:
1.1 1 x 1 real = 5
1.2 1 x 1 real = 5
The problem is, that the first value of return vector is overwritten
by the value of the second element. We do not understand why?
Did we miss something obvious?
Many regards
Uli (and Magdalena Luniak)
*
* 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/
