Tom/Laurel,
Thanks for your helpful replies.
Tom,
And, yup...gulp...blush, I managed to confuse the incidence (Po) stat with
the p-value. Hopefully, I am awake now.
Thanks
----- Original Message -----
From: "Steichen, Thomas" <[email protected]>
To: <[email protected]>
Sent: Wednesday, July 16, 2003 5:08 PM
Subject: st: RE: relative risk
> I don't see what you are referring to, Wendell. The p-values
> (labeled "P>|Z|") appear to be changing to me. What did not
> change is Po, which is defined as "the incidence of the outcome
> (dependent variable = 1) in the unexposed group." This should not
> change based on # of predictors in the model.
>
> Tom
>I know nothing about this command, but it is probably just your large
sample
>size that is driving your overall results (Prob > chi2 = 0.0000 reports
>on the model). Your P>|z| for individual predictors varies.
>To check, take a 1% subsample of your dataset and run the same analysis.
With 180 or 190 obs, you might get a less sig model.
Laurel
Ann Arbor VAMC
>
> > -----Original Message-----
> > From: Wendell Joice [mailto:[email protected]]
> > Sent: Wednesday, July 16, 2003 4:55 PM
> > To: [email protected]
> > Subject: st: relative risk
> >
> >
> > I am using Stata 7 and the -relrisk- (ado) command in conjunction
> > ith -logistic-. I note that, for my data, the p value (for a given
> > covariate) never changes, regardless of how many or which
> > variables I add to
> > the model. For example, in the data below, I am using a
> > dependent variable
> > (psyment) with a covariate (racewbh). The relative risk ratios for
> > _Iracewbh_2 & _Iracewbh_3 change as I add other covariates
> > to the model;
> > but the p value for these two variables never changes (see
> > small and large
> > models below).
> >
> > Is this the way it should be or am I in error with my data?
> > My usage of
> > the -relrisk- command?
> > Is there a simple way to check (verify) the output of the
> > -relrisk- command?
> >
> > Thanks,
> > Wendell Joice
> >
> >
> >
> > SMALL MODEL
> >
> > xi: logistic psyment i.racewbh if csample==1
> > i.racewbh _Iracewbh_1-3 (naturally coded;
> > _Iracewbh_1
> > omitted)
> >
> > Logit estimates
> > Number of obs =
> > 19295
> > LR
> > chi2(2) =
> > 68.54
> > Prob
> > > chi2 =
> > 0.0000
> > Log likelihood = -12961.957
> > Pseudo R2 =
> > 0.0026
> >
> >
> > ----------------------------------------------------------------------
> > --------
> > psyment | Odds Ratio Std. Err. z P>|z|
> > [95% Conf.
> > Interval]
> >
> > -------------+--------------------------------------------------------
> > --------
> > _Iracewbh_2 | .6429203 .0385832 -7.36 0.000
> > .5715766
> > .723169
> > _Iracewbh_3 | .7907062 .0466184 -3.98 0.000
> > .7044174
> > .8875651
> >
> > ----------------------------------------------------------------------
> > --------
> >
> > relrisk
> > Po Rel Risk
> > Lower
> > Limit Upper Limit
> > RR _Iracewbh_2: 0.066 0.658 0.588
> > 0.737
> > RR _Iracewbh_3: 0.065 0.802 0.718
> > 0.894
> >
> >
> > LARGER MODEL BELOW
> >
> > xi: logistic psyment i.racewbh i.physpecy i.psydin2
> > i.phytimec i.chrcadg
> > i.di
> > > agsrvc i.acutcadg i.newinsur i.likreccdg i.psydrnumx
> > i.agecatc
> > i.GEOGREGN i.
> > > yearmhp if csample==1
> >
> > (output abbreviated)
> >
> > Logit estimates
> > Number of obs =
> > 18110
> > LR
> > chi2(30) =
> > 11484.07
> > Prob
> > > chi2 =
> > 0.0000
> > Log likelihood = -6465.4055
> > Pseudo R2 =
> > 0.4704
> >
> >
> > ----------------------------------------------------------------------
> > --------
> > psyment | Odds Ratio Std. Err. z P>|z|
> > [95% Conf.
> > Interval]
> >
> > -------------+--------------------------------------------------------
> > --------
> > _Iracewbh_2 | .8734043 .0805943 -1.47 0.142
> > .7289031
> > 1.046552
> > _Iracewbh_3 | .9309302 .0823622 -0.81 0.419
> > .7827242
> > 1.107199
> > _Iphyspecy_2 | .5840736 .0408614 -7.69 0.000
> > .5092348
> > .6699109
> > _Iphyspecy_3 | 10.01045 .8514816 27.08 0.000
> > 8.473265
> > 11.82649
> > ..........................................Output
> > abbreviated......................
> >
> > _Iyearmhp_6 | .4989542 .0376058 -9.22 0.000
> > .4304337
> > .5783824
> >
> > ----------------------------------------------------------------------
> > --------
> >
> > relrisk
> > Po Rel Risk
> > Lower
> > Limit Upper Limit
> > RR _Iracewbh_2: 0.066 0.881 0.742
> > 1.043
> > RR _Iracewbh_3: 0.065 0.935 0.794
> > 1.100
> > RR _Iphyspecy_2: 0.168 0.628 0.555
> > 0.709
> > RR _Iphyspecy_3: 0.016 8.764 7.579
> > 10.100
> > ............................................Output
> > abbreviated...............................
> > RR _Iyearmhp_5: 0.059 0.773 0.667
> > 0.896
> > RR _Iyearmhp_6: 0.061 0.515 0.446
> > 0.594
> >
> >
> >
> > *
> > * 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/
> >
>
> -----------------------------------------
> CONFIDENTIALITY NOTE: This e-mail message, including any
> attachment(s), contains information that may be confidential,
> protected by the attorney-client or other legal privileges, and/or
> proprietary non-public information. If you are not an intended
> recipient of this message or an authorized assistant to an intended
> recipient, please notify the sender by replying to this message and
> then delete it from your system. Use, dissemination, distribution,
> or reproduction of this message and/or any of its attachments (if
> any) by unintended recipients is not authorized and may be unlawful.
>
>
> *
> * 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/
