Bookmark and Share

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: can I use -parmest- with -mlogit-?


From   Roger Newson <[email protected]>
To   "[email protected]" <[email protected]>
Subject   Re: st: can I use -parmest- with -mlogit-?
Date   Mon, 14 Jun 2010 17:52:18 +0100

Jamie must have had a very old -parmest- before updating it to Stata Version 10. The requirement for at least 1 output-destination option came in with the update dated 10 November 2002, when we still had Stata Version 7. As I recall, the change was urged by Bill Gould at the 2002 UK Stata User Meeting (20-21 May, 2002), because most Stata programs try to protect users from overwriting their original data except if they specifically ask Stata to do this. Fortunately, Jamie now seems to have the most recent Stata 10 version (dated 22 April 2009).

Best wishes

Roger


Roger B Newson BSc MSc DPhil
Lecturer in Medical Statistics
Respiratory Epidemiology and Public Health Group
National Heart and Lung Institute
Imperial College London
Royal Brompton Campus
Room 33, Emmanuel Kaye Building
1B Manresa Road
London SW3 6LR
UNITED KINGDOM
Tel: +44 (0)20 7352 8121 ext 3381
Fax: +44 (0)20 7351 8322
Email: [email protected]
Web page: http://www.imperial.ac.uk/nhli/r.newson/
Departmental Web page:
http://www1.imperial.ac.uk/medicine/about/divisions/nhli/respiration/popgenetics/reph/

Opinions expressed are those of the author, not of the institution.

On 14/06/2010 17:21, Jamie Fagg wrote:
Thanks to Roger and Maarten, your replies were very helpful.

I meant to reply this morning, but the message was returned by the Major.

So, Maarten said that "it seems to me that the problem is that you are
using an old version of -parmest-. This is why it is crucial to
specify where you found a user written program: there are often
multiple version floating around in cyber space. Often the one on SSC
is the most recent."

Thanks for this advice Maarten. I'll try to do that in future when
posting about user-written commands.

and Roger added that "parmest (without any options) should fail after
ANY estimation command, because  -parmest- requires at least 1 of the
output-destination options -list()-,
   >  -saving()-, -norestore- and/or -fast-."

Interestingly, the command didn't fail when I ran it with no options
with logit, but seemed to do what I required (albeit writing over the
dataset in memory when it did so), but I shall make sure that I use
the options once I've got it up and running.

Thanks also to Roger, for your subsequent follow up with regards to
the 10.1 version which I was just about to flag. I've got the 10.1
version installed and it seems to be working fine - I can now make my
graph!

Best wishes,

Jamie

On 14 June 2010 11:56, Jamie Fagg<[email protected]>  wrote:
Dear Roger and Maarten,

Thanks for your replies, all of which have been very helpful.

Maarten said:


So, it seems to me that the problem is that you are using an
old version of -parmest-. This is why it is crucial to
specify where you found a user written program: there are
often multiple version floating around in cyber space. Often
the one on SSC is the most recent.

Thanks for this advice Maarten. I'll try to do that in future when posting
about user-written commands.

Roger said:

parmest

(without any options) should fail after ANY estimation command, because
-parmest- requires at least 1 of the output-destination options
-list()-,
-saving()-, -norestore- and/or -fast-.

Interestingly, the command didn't fail, but seemed to do what I required
(albeit writing over the dataset in memory when it did so), but I shall make
sure that I use the options once I've got it up and running.

Thanks also for your subsequent follow up with regards to the 10.1 version
which I was just about to flag. I've got the 10.1 version installed and it
seems to be working fine - I can now make my graph!

Best wishes,

Jamie

On 14 June 2010 11:40, Roger Newson<[email protected]>  wrote:
Yes, you can use -parmest- after -mlogit-, and I do this routinely.
However,
the command

parmest

(without any options) should fail after ANY estimation command, because
-parmest- requires at least 1 of the output-destination options -list()-,
-saving()-, -norestore- and/or -fast-. These output-destination options
control whether the output dataset is listed, saved to a disk file, or
written to the memory, overwriting the original data. (These options are
not
mutually exclusive.)

I tried to replicate something similar to Jamie's example with -mlogit-,
by
typing

use http://www.stata-press.com/data/lf2/ordwarm2,clear
mlogit warm yr89,nolog
parmest, list(, sepby(eq))

and this worked for me (under Stata 11.1), producing a listing of the
parameters, separated by the equation variable -eq-, with P-values and
confidence limits formatted more informatively than as provided by
-mlogit-.
I could equally have used the -saving()-, -norestore- or -fast- options,
if
I wanted to save the results to disk or to the memory, respectively.

I hope this helps.

Best wishes

Roger



Roger B Newson BSc MSc DPhil
Lecturer in Medical Statistics
Respiratory Epidemiology and Public Health Group
National Heart and Lung Institute
Imperial College London
Royal Brompton Campus
Room 33, Emmanuel Kaye Building
1B Manresa Road
London SW3 6LR
UNITED KINGDOM
Tel: +44 (0)20 7352 8121 ext 3381
Fax: +44 (0)20 7351 8322
Email: [email protected]
Web page: http://www.imperial.ac.uk/nhli/r.newson/
Departmental Web page:

http://www1.imperial.ac.uk/medicine/about/divisions/nhli/respiration/popgenetics/reph/

Opinions expressed are those of the author, not of the institution.

On 14/06/2010 11:07, Jamie Fagg wrote:

Dear all,

I am trying to write estimates from an -mlogit- command using the
-parmest- by Roger Newson. He seems to suggest that this should be
possible (see
http://www.stata.com/statalist/archive/2002-12/msg00025.html)

Parmest seems to work fine (using Stata 10.1) when I try it with logit
(see below), but returns an error when run with mlogit. This
replicates the error that I find when I try to use it with my own
data. I can't find any specific references to this in the parmest help
file. Do I need to specify further options to use parmest with mlogit?

Thanks,

Jamie

. use http://www.stata-press.com/data/lf2/ordwarm2,clear
(77&   89 General Social Survey)

. logit white age,nolog

Logistic regression                               Number of obs   =
2293
                                                  LR chi2(1)      =
5.84
                                                  Prob>   chi2     =
0.0157
Log likelihood = -853.93381                       Pseudo R2       =
0.0034



------------------------------------------------------------------------------
       white |      Coef.   Std. Err.      z    P>|z|     [95% Conf.
Interval]


-------------+----------------------------------------------------------------
         age |   .0093244    .003902     2.39   0.017     .0016767
  .0169721
       _cons |   1.550548   .1794133     8.64   0.000     1.198904
  1.902191


------------------------------------------------------------------------------

. parmest

. use http://www.stata-press.com/data/lf2/ordwarm2,clear
(77&   89 General Social Survey)

. mlogit warm yr89,nolog

Multinomial logistic regression                   Number of obs   =
2293
                                                  LR chi2(3)      =
  76.96
                                                  Prob>   chi2     =
0.0000
Log likelihood = -2957.2879                       Pseudo R2       =
0.0128



------------------------------------------------------------------------------
        warm |      Coef.   Std. Err.      z    P>|z|     [95% Conf.
Interval]


-------------+----------------------------------------------------------------
SD           |
        yr89 |  -1.174919   .1591851    -7.38   0.000    -1.486916
  -.862922
       _cons |  -.6803538    .079895    -8.52   0.000    -.8369451
-.5237625


-------------+----------------------------------------------------------------
D            |
        yr89 |  -.3810336   .1033743    -3.69   0.000    -.5836435
-.1784237
       _cons |  -.0129591   .0657255    -0.20   0.844    -.1417787
  .1158604


-------------+----------------------------------------------------------------
SA           |
        yr89 |   .0964589   .1196593     0.81   0.420     -.138069
  .3309868
       _cons |  -.7642883    .082184    -9.30   0.000    -.9253661
-.6032105


------------------------------------------------------------------------------
(warm==A is the base outcome)

. parmest
Data will not be replaced
Estimates and SEs could not be extracted
r(498);

*
*   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/




--
Dept. of Geography, Queen Mary, University of London
Mile End Rd
E1 4NS

Tel: 020 7882 5400







--
Dept. of Geography, Queen Mary, University of London
Mile End Rd
E1 4NS

Tel: 020 7882 5400

*
*   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/


© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index