Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: RE: RE: -forval- with -inequal7-


From   philippe van kerm <philippe.vankerm@ceps.lu>
To   "statalist@hsphsun2.harvard.edu" <statalist@hsphsun2.harvard.edu>
Subject   st: RE: RE: -forval- with -inequal7-
Date   Tue, 15 Jul 2008 13:43:41 +0200

Lola,

-inequal7- exits with an error message when run on a (sub-)sample of 0 observations. Some commands do work in this case (e.g. summarize), others don't (e.g. regress).

A simple workaround here is to -capture- the output of -inequal7-

g gini=.
qui forval i = 19901/20071 {
  capture inequal7 income [fw=wei] if date==`i', returns
  replace gini=r(gini) if date==`i'
}

But it would probably be cleaner to loop only over valid 'dates'. See the FAQ suggested by Nick Cox: http://www.stata.com/support/faqs/data/foreach.html

Philippe

> -----Original Message-----
> From: owner-statalist@hsphsun2.harvard.edu [mailto:owner-
> statalist@hsphsun2.harvard.edu] On Behalf Of Lola Jackson
> Sent: Tuesday, July 15, 2008 11:09 AM
> To: statalist@hsphsun2.harvard.edu
> Subject: st: RE: -forval- with -inequal7-
>
>
> (I've been trying to send this message since yeterday but it's not
> appearing on the statalist, hope it does now...)
>
>
>
> Apologies for some typos in my previous query, I copied and pasted my
> code into my message and then modified it for the various commands but
> mixed them up in the email.. However, this is not the problem. An error
> message arises whenever there is a 'missing' date, using -inequal7-.
>
> The codes I have tried are as follows:
>
> using forval:
>
> g gini=.
> qui forval i = 19901/20071 {
> inequal7 income [fw=wei] if date==`i', returns
> replace gini=r(gini) if date==`i'
> }
>
> and using foreach:
>
> g gini=.
> qui foreach i of num 19901/20071 {
> inequal7 income [fw=wei] if date==`i', returns
> replace gini=r(gini) if date==`i'
> }
>
> But with both get the error message:
> '0' invalid obs no
> r(198);
>
> These do work when limited to 'consecutive' dates (eg 19901/19904), the
> problem arises when they span  'missing' dates as there is no 19905.
> The programme runs and calculates the gini for the dates up until a
> 'missing' date and then stops with an error. This is not a problem with
> other commands using the same dataset, but has arisen with -inequal7-.
>
> Thanks,
> Lola
>
>
>
>
> --- On Mon, 7/14/08, Nick Cox <n.j.cox@durham.ac.uk>
> wrote:
>
> > From: Nick Cox <n.j.cox@durham.ac.uk>
> > Subject: st: RE: -forval- with -inequal7-
> > To: statalist@hsphsun2.harvard.edu
> > Date: Monday, July 14, 2008, 5:44 PM
> > -inequal7- is blameless here.
> >
> > Your -forval- syntax, at least as given here, is
> incorrect.
> >
> >
> > forval i = 19901/20071 {
> >
> > You must have an equals sign. See the help on
> -forvalues-,
> > including all
> > the examples given.
> >
> > Similarly, your first example with -foreach- has
> incorrect
> > syntax.
> >
> > You cannot mix and match the two syntaxes. Also,
> guessing
> > at syntax you
> > would like or imagine might work is all too likely just
> to
> > cause
> > frustration.
> >
> > Given the gaps, you might be better advised to check
> out
> > solutions
> > documented at
> >
> > FAQ     . . . . .
> . . . . . Making foreach go through all
> > values of a
> > variable
> >         8/05    Is there a way to tell Stata to try all
> > values of a
> >                 particular variable in a foreach statement
> > without
> >                 specifying them?
> >
> > http://www.stata.com/support/faqs/data/foreach.html
> >
> > Nick
> > n.j.cox@durham.ac.uk
> >
> > Lola Jackson
> >
> > I am using -forval- to run -inequal7- by date and
> export
> > selected
> > results. The problem is that I get the error message:
> >
> > '0' invalid obs no
> > r(198);
> >
> > I think the reason is that by dates are not
> > 'consecutive' in that I have
> > 19901, 19902, 19903, 19904, then 19911, 19912, etc. (As
> > there are
> > quarterly observations for each year 1990, 1991 etc).
> From
> > various tests
> > I have concluded that the error arises when the loop
> tries
> > to run for a
> > date when there are no observations (eg 19905).
> >
> > My code is:
> > qui forval i 19901/20071 {
> > inequal7 income [fw=wei] if date==`i', returns
> > replace gini=r(gini) if date==`i'
> > }
> >
> > I also tried using -foreach- but got the same error
> > message:
> > qui foreach i = 19901/20071 {
> > inequal7 income [fw=wei] if date==`i', returns
> > replace gini=r(gini) if date==`i'
> > }
> >
> > The only way that it works is if I specify each date
> > individually, as
> > in:
> > qui foreach i in 19901 19902 19903 **etc** {
> > inequal7 income [fw=wei] if date==`i', returns
> > replace gini=r(gini) if date==`i'
> > }
> >
> > But this means typing each date individually.
> >
> > I have run -forval- in this way for the same dataset,
> > without problems,
> > the problem seems to arise when combined with
> -inequal7-.
> >
> > I would welcome any suggestions as to how to correct
> the
> > problem (or
> > even a smarter way of doing what I need!)
> >
> >
> > *
> > *   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/
>
> ________________________________
> Not happy with your email address?
> Get the one you  really want - millions of new email addresses
> available now at Yahoo!
>
>
>       __________________________________________________________
> Not happy with your email address?.
> Get the one you really want - millions of new email addresses available
> now at Yahoo! http://uk.docs.yahoo.com/ymail/new.html
>
>
>
> *
> *   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/

*
*   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/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index