RE: st: Count observations

["Impavido, Gregorio" <GImpavido@imf.org>]

I am now unclear what you want to count.... so I try few things.

The following counts the number of times either A or B  appears in your variable _over all ids_ (I am following your statement that "A" and "B" are the "same" or "equivalent")

******* start example
count if fee=="A" | fee=="B"
******* end example

This does it _within each id_ (you had worked it out yourself)

******* start example
bysort id : count if fee=="A" | fee=="B"
******* end example

There are other things you can do. For instance, # of firms with either A or B, # of firms with both A and B, et cetera. In your second email you appear to be interested in the # of firms that have either A or B.  This can be done by:

******* start example
tab id if fee=="A" | fee=="B"
di r(r)
******* end example

while this counts the number of firms that have both "A" and "B" (but this crucially assumes that both "A" and "B", if they appear, cannot appear more than once. If either "A" or "B" can appear more than once by id, it does not work)

******* start example
gen touse = (fee=="A" | fee=="B")
bysort id : egen total = total(touse)
tab id if total==2
di r(r)
******* end example



-----Original Message-----
From: owner-statalist@hsphsun2.harvard.edu [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Miguel Angel Duran Munoz
Sent: Tuesday, July 23, 2013 12:07 PM
To: statalist@hsphsun2.harvard.edu
Subject: RE: st: Count observations

Thank you very much for your help. Let me explain a bit more why -count- did not work. There is something in my variables that I did not make explicit in my first message (I thought could solve it on my own after being helped, but it is not the case).

As I told you, the variable fee describes the type of fee (eg, A B C).
Nevertheless, the dataset is constructed in a way that A and B, for instance, are the same (specifically, I have "commitment fee" and "commitment regular fee", but both types are the same). But, although A and B are the same, they both might be included for the same firm.

Therefore, given this illustrative dataset,

Id    Type-of-fee

1          A
1          B
1          C
2          C
2          A
3          A
4          B
4          .
4          A

there are 4 firms that have either A or B. I was trying to use this, -bysort id: count if fee=="A" | fee=="B", but what I get is (obsviously) split by firms.

I am sorry for the initial confusion.

Miguel.

 Unclear why it does not work. It works with the following:
>
> ******* start example
> clear all
> input id
> 1
> 1
> 1
> 2
> 2
> 3
> 4
> 4
> 4
> end
> input str2 fee
> A
> B
> C
> C
> A
> A
> B
> ""
> A
> count if fee=="A"
> ******* end example
>
> Notice that another alternative is -tab fee-
>
> -----Original Message-----
> From: owner-statalist@hsphsun2.harvard.edu
> [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Miguel 
> Angel Duran Munoz
> Sent: Tuesday, July 23, 2013 10:51 AM
> To: statalist@hsphsun2.harvard.edu
> Subject: Re: st: Count observations
>
> Hi, Statalisters. I have the following doubt. My dataset is arranged 
> in the following way. I have a variable that identifies firms (say id).
> Another variable describes whether different types of fees (eg, A B C) 
> applies to a firm. Accordingly, the dataset looks similar to
>
> Id    Type-of-fee
>
> 1          A
> 1          B
> 1          C
> 2          C
> 2          A
> 3          A
> 4          B
> 4          .
> 4          A
>
> I would like to know, for instance, the number of A fees that there 
> are. I have used -count- but I am not able to get what I want. Will 
> you please help me?
>
> Thanks in advance.
>
> Miguel.
>
>
> *
> *   For searches and help try:
> *   http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/faqs/resources/statalist-faq/
> *   http://www.ats.ucla.edu/stat/stata/
>
> *
> *   For searches and help try:
> *   http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/faqs/resources/statalist-faq/
> *   http://www.ats.ucla.edu/stat/stata/
>


*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/