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: Concord - plottting problem


From   Martyn Sherriff <statams48@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Concord - plottting problem
Date   Tue, 7 May 2013 08:56:48 +0100

Thank you for the work-around, and the comments about the range which
are appreciated.

Martyn

On 7 May 2013 08:49, Nick Cox <njcoxstata@gmail.com> wrote:
> That behaviour is wired into the code at present, although it could be changed.
>
> I suppose showing the entire range could be justified as a feature
> allowing  consistent comparison, but I have absolutely no recall of
> whether that's my code or Thomas Steichen's or of whether it was
> deliberate or an oversight. (-concord-'s first public appearance was
> 1998.) TJS = Thomas J. Steichen, twice retired, NJC = Nicholas J. Cox,
> still responsible for this child.
>
> There's a work-around.
>
> levelsof site, local(sites)
>
> foreach s of local sites {
>         preserve
>         keep if site == `s'
>         concord <whatever>
>         restore
> }
>
> A completely different comment is that with values like those given,
> ranging from about 14 to 370, multiplicative comparisons may be in
> order, i.e. compare ln(x1) and ln(x2) as well.
>
>
> Nick
> njcoxstata@gmail.com
>
>
> On 7 May 2013 07:36, Martyn Sherriff <statams48@gmail.com> wrote:
>> I am using concord (package concord from
>> http://fmwww.bc.edu/repec/bocode/c, 3.1.0 10nov2010  TJS & NJC
>> graphics improvements ) to look at agreement. I have several sites to
>> look at, so I stacked the data as x1, x2 and site as a variable to
>> select the data. When I use the plotting options of 'loa' and 'qnormd'
>> selecting on the value of site the plots are satisfactory. However
>> when I use the option 'ccc', the one that I am interested in, the axes
>> take the maximum values of the entire dataset ignoring the 'if'
>> condition. I have tried to limit the axes to no avail:
>>
>> concord x1 x2 if site == 1,loa
>> results in an x-axis upper limit of 100, whereas
>> concord x1 x2 if site == 1,ccc
>> results in an x-axis upper limit of 400
>>
>> The summary data is:
>> . tabstat x1 x2,stats(n min max) by(site)
>>
>> Summary statistics: N, min, max
>>   by categories of: site
>>
>>     site |        x1        x2
>> ---------+--------------------
>>        1 |        13        13
>>          |     13.82     13.82
>>          |     98.36      98.4
>> ---------+--------------------
>>        2 |        17        17
>>          |     37.77     38.25
>>          |       360       369
>> ---------+--------------------
>>    Total |        30        30
>>          |     13.82     13.82
>>          |       360       369
>> ------------------------------
>>
>> I realise that I am missing something quite simple in the syntax but I
>> just cannot see it.
> *
> *   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/


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