Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
RE: st: Setting confidence intervals with -prvalue- after -ologit-
From
"MacLennan, Graeme" <[email protected]>
To
"'[email protected]'" <[email protected]>
Subject
RE: st: Setting confidence intervals with -prvalue- after -ologit-
Date
Thu, 1 Sep 2011 16:19:53 +0100
Of course! OK, so -trace- is useful, reading its help file, something new learned. Thanks. G.
-----Original Message-----
From: [email protected] [mailto:[email protected]] On Behalf Of Nick Cox
Sent: 01 September 2011 16:12
To: [email protected]
Subject: Re: st: Setting confidence intervals with -prvalue- after -ologit-
Good question.
I started with
. set trace on
. set traced 1
and ran your example. I think I had to bump up to -set traced 2-.
This way I found that -_pepred- was complaining and checked that the problem lay in your call being rejected by its -syntax- statement.
This looked like a helper program but a check showed that it has its own help file.
So it's easy!
Nick
On Thu, Sep 1, 2011 at 3:55 PM, MacLennan, Graeme <[email protected]> wrote:
> Nick thank you for this information. As an aside, I would like to try and track information like this down myself in future. How did you know to look "there"?
>
>
>
> -----Original Message-----
> From: [email protected]
> [mailto:[email protected]] On Behalf Of Nick Cox
> Sent: 01 September 2011 15:40
> To: [email protected]
> Subject: Re: st: Setting confidence intervals with -prvalue- after
> -ologit-
>
> _pepred in this package insists on integer arguments; that is what biting. The restriction is documented, but you need to look at the specific help file. I don't know of any arguments for this except convention, but you need to ask the authors.
>
> Nick
>
> On Thu, Sep 1, 2011 at 3:22 PM, MacLennan, Graeme <[email protected]> wrote:
>> I am using -prvalue- from -spost- version 9 from http://www.indiana.edu/~jslsoc/stata (-findit spost-) to get predicted probabilities and differences after using -ologit-. The command -prvalue- has a -level(#)- option to specify the width of confidence intervals. I wanted to use 97.5% Cis but it would appear that only integer values are accepted? The following syntax tells me that "option level() incorrectly specified" and "invalid syntax" repectively. If this can be reproduced, I will send off an email to the address in the help file. (Perhaps there is a good reason, or I am doing something stupid).
>> Graeme.
>>
>>
>> sysuse auto, clear
>> ologit rep78 price foreign, or level(97.5) prvalue, x(foreign=0)
>> level(97.5)
>>
>> set level 97.5
>> ologit rep78 price foreign, or
>> prvalue, x(foreign=0)
>>
>
> *
> * 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/
>
>
> The University of Aberdeen is a charity registered in Scotland, No SC013683.
>
> *
> * 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/
>
*
* 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/
The University of Aberdeen is a charity registered in Scotland, No SC013683.
*
* 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/