[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

# st: relative risk

 From "Wendell Joice" <[email protected]> To <[email protected]> Subject st: relative risk Date Wed, 16 Jul 2003 16:55:15 -0400

```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/
```

 © Copyright 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index