Re: st: RE: 'if' with labels but finds no observations - sorted!

["Paul O'Brien" <paul.obrien@afe2.org.uk>]

The ncyear (year_f) has the value 2003 and the value label 2003/4

So for the financial year from 1st April 2003

. table ageint if year_f==2003

----------------------
ageint |      Freq.
----------+-----------
10 |          1
11 |          1
12 |          7
13 |         12
14 |         99
15 |        325
16 |        700
17 |        960
18 |      1,085
19 |      1,002
20 |      1,213
21 |      1,348


On 23 Apr 2004, at 19:07, Nick Cox wrote:

> There is no reason to expect -if- to accept
> value labels in this way. Unless explicitly 
> instructed, -if- deals with values, not 
> value labels. Thus 
>
> if ncyear == 2000/1 
>
> is being interpreted as 
>
> if ncyear == 2000 
>
> as the expression 2000/1 evaluates
> to 2000. On the other hand you evidently 
> have no years that are 2001/2, i.e. 1000.5. 
>
> The syntax is 
>
> if <exp> 
>
> where <exp> denotes an expression, which will 
> be evaluated before being used. 
>
> On the other hand, something like this 
> should work: 
>
> tab ncyear if ncyear == "2000/1":name
>
> Nick 
> n.j.cox@durham.ac.uk 
>
> Paul O'Brien
>
> > I am not sure what the problem is here, but 'if' is not working for   
> > some date labels.
> > year_f is OK up to 2000/1 but beyond that give no observations.
> >
> > .egen year_f = ncyear(date_v), m(4)
> >
> > . table year_f
> > ----------------------
> > ��year_f |�����Freq.
> > ----------+-----------
> > �1997/98 |����44,023
> > �1998/99 |����42,654
> > ��1999/0 |����42,944
> > ��2000/1 |����42,127
> > ��2001/2 |����41,520
> > ��2002/3 |����41,463
> > ��2003/4 |����40,057
> > ----------------------
> >
> > . table ageint if year_f==2000/1
> >
> > ----------------------
> > ��ageint |�����Freq.
> > ----------+-----------
> > ������12 |���������5
> > ������13 |��������36
> > ������14 |�������131
> > ������15 |�������346
> > etc
> >
> > . table ageint if year_f==2001/2
> >
> > no observations
> >
> > r(2000);
> >
> > But there are observations for year_f==2001/2
> >
> > �. table ageint year_f
> >
> > --------------------------------------------------------------
> > ---------- 
> > -
> > |                            year_f
> > ageint | 1997/98  1998/99   1999/0   2000/1   2001/2   2002/3    
> > 2003/4
> > ---------- 
> > +--------------------------------------------------------------
> > 10 |       1        1                          2      
> > 1         
> > 1
> > 11 |       2        2        2                 1      
> > 2         
> > 1
> > 12 |       2        3        2        5        4      
> > 3         
> > 7
> > 13 |      13       20       17       36       22      
> > 20        
> > 12
> > 14 |      73       49       64      131      110      
> > 101        
> > 99
> > 15 |     238      259      238      346      362      
> > 302       
> > 325
> >
> > If I try -list- I get the same thing:
> >
> > . list if year_f==2000/1
> >
> > �Observation 129622
> >
> > ���clin_s~e����������RC���clin_w~k������������������connum����
> > �������2�� 
> > �����date���15-Feb-01��������dob���04-Apr-63
> >
> > etc
> >
> > �
> > Observation 129623
> >
> > ���clin_s~e����������RC���clin_w~k�������D- 
> > CON�����connum�����������1�������date���07-Feb-01��������dob���06-Apr 
> > -77
> >
> > etc
> >
> > . list if year_f==2001/2
> >
> > This last command gives nothing.
>
> *
> *   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/

Dr Paul O'Brien
Westside Contraceptive Services
Raymede Clinic, Exmoor St, London
W10 6DZ
020 8962 4450 - tel
020 8962 4451 - fax
07900 495 118 - mob