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, statalist.org is already up and running.


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

Re: st: RE: AW: RE: AW: Identifying unique values with codebook


From   Steve Samuels <sjsamuels@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: RE: AW: RE: AW: Identifying unique values with codebook
Date   Wed, 16 Jun 2010 10:39:26 -0400

I agree. I was incorrect in stating that --inspect- and -summarize-
treat strings alike. -summarize- states that there are no observations
on a string variable. -inspect- does count the number of observations,
but then reports nonsense.

Steve

On Wed, Jun 16, 2010 at 10:31 AM, Nick Cox <n.j.cox@durham.ac.uk> wrote:
> -inspect- has its fans, but I am not among them.
>
> -inspect- seems to regard strings as failed numerics, and refuses to
> look inside. It would be simpler to ignore all strings.
>
> I wonder whether some kind of review of -inspect- is called for.
> Whatever -inspect- does that differs from -codebook- could be taken into
> -codebook-, perhaps. -inspect- could remain available under version
> control.
>
> Nick
> n.j.cox@durham.ac.uk
>
> Martin Weiss
>
> A "string representation" seems to work with -codebook-, but not
> -inspect-
> which claims to find o unique values...
>
>
> *************
> clear*
> inp str25 mystrvar
> loooooooooooooooooooooooooooonnggg
> verylooooooooooooooooooooonnnnnnnnngggg
> end
> ins
> codebook
> *************
>
> Nick Cox
>
> As Martin says, at root this is a precision problem.
>
> Neither -codebook- nor anything else is to blame if it is presented with
> the same values. To hold very large integers you may need to consider
> -long- as another possibility, or even a string representation.
>
> Martin Weiss
>
> As -help data_types- says: "doubles have 16 digits of accuracy." So you
> can
> increase the digits of your "y" up to the point where even -double- can
> do
> nothing for you:
>
> *************
> clear
> set obs 10
> gen byte x=_n
> codebook x
> gen double y1 = 1000000000000000 + x
> gen double y2 = 10000000000000000 + x
> gen double y3 = 100000000000000000 + x
> gen double y4 = 1000000000000000000 + x
> codebook y?
> ins y?
> *************
>
>
> Interestingly, -inspect- seems to differ from -codebook-`s opinion.
>
> Walter Garcia-Fontes
>
> I stumped into a problem when identifying unique values of a numeric
> variable using "codebook": if the values are large they will be
> identified as the same value.
>
> For instance I have a variable x with the following values:
> 0, 1, 2, ... 10 (that is 10 different values)
>
> codebook x
> reports "unique values: 10"
>
> Now do
> gen y = 100000000000000000 + x
>
> codebook y
> reports "unique values = 1"
>
> Is this a feature?
>
> *
> *   For searches and help try:
> *   http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/statalist/faq
> *   http://www.ats.ucla.edu/stat/stata/
>



-- 
Steven Samuels
sjsamuels@gmail.com
18 Cantine's Island
Saugerties NY 12477
USA
Voice: 845-246-0774
Fax:    206-202-4783

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   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   |   Site index