Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, is already up and running.

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: loop

From   Nick Cox <>
To   "" <>
Subject   Re: st: loop
Date   Mon, 27 Aug 2012 08:46:11 +0100

It's not possible to comment on Chiara's code or what went wrong because contrary to much repeated advice there are no precise details of what she tried.

But given say -newcat121-newcat1212- the log of the frequencies of 1s are given by

forval j = 1/12 {
       su newcat12'j',  meanonly
       .... = ln(r(sum))

where ... is the assignment to scalar or variable Chiara wants.

But the dummies (indicators) are not necessary for this purpose at all. A -tabulate- of the original variable shows the frequencies in question.

(For ' read left single quote above when appropriate.)


On 27 Aug 2012, at 08:22, Muhammad Anees <> wrote:

tab cat12, gen(newcat12)

will give you those required variables (categorical or dummies).


On Mon, Aug 27, 2012 at 11:32 AM, Chiara Mussida <> wrote:
Dear all,
the variable cat12 takes the values from 1 to 12. I want to create 12
different scalars or variables for each cat. The problem is that each
scalar must take the values correspondent to the log of the number of
obs in each cat (weighted) since i have to write an index which needs
those numbers. I tried with forvalues but i did not succeed!
Ps: cat12 is a variable for occupational categories (from 1 to 12). In
the index i put the weighted values (weight available in my dataset).

*   For searches and help try:

© Copyright 1996–2016 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index