I don't understand this enough to answer you.
I see that some macro `nj' must be defined
upstream of this.
Either this is the whole problem and you
have simple calculations involving the numbers
10, 20, 30, 40, which wouldn't seem to
need a program. You should enter those
numbers as data.
Or else you are steadily generalising this,
and 10, 20, 30, 40 are just placeholders,
in which case it still sounds as if you have some
standard problem for which Stata commands
already exist.
As said, I don't understand this, so I may
be wide of the mark.
Nick
[email protected]
Serguei Kaniovski
> OK, I have corrected the first two errors, but how to compute
> sum11_i_j, sum10_i_j, etc over sumT_i_j in the second loop,
> i.e. instead
> of ..., which should be a scalar of high precision. Thanks.
>
> *INITIALIZE SUMS;
> forval i=1/5 {;
> local I=`i'+1;
> forval j=`I'/`nj' {;
> local sum11_`i'_`j'=10;
> local sum10_`i'_`j'=20;
> local sum01_`i'_`j'=30;
> local sum00_`i'_`j'=40;
> local sumT_`i'_`j'=100;
> };
> };
>
> *COMPUTE RELATIVE FREQUENCIES;
> forval i=1/5 {;
> local I=`i'+1;
> forval j=`I'/`nj' {;
> local frq11_`i'_`j'=...;
> local frq10_`i'_`j'=...;
> local frq01_`i'_`j'=...;
> local frq00_`i'_`j'=...;
> };
> };
*
* 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/
