Dear All,
I have a very practical question
I am running a multivariate model from which i would like to delete
the non relevant variables. The SVY commands don t support LRT so the
-test- would be the solution but the -test- is for two variables at a
time? How to proceed with 12?.
Thanks
--------------------------------------------------------------------------
. svy: logit sd106f v106 v201 v013 v025 v190 s708 s109 v513 v024
v130 s801 v743d v731 ,or
(running logit on estimation sample)
Survey: Logistic regression
Number of strata = 34 Number of obs = 1529
Number of PSUs = 1449 Population size = 1674.0756
Design df = 1415
F( 13, 1403) = 7.50
Prob > F = 0.0000
------------------------------------------------------------------------------
| Linearized
sd106f | Odds Ratio Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
v106 | .859833 .0807522 -1.61 0.108 .7151615 1.03377
v201 | .8916687 .0880231 -1.16 0.246 .7346885 1.082191
v013 | 1.045695 .1223353 0.38 0.703 .8312629 1.315443
v025 | 1.047904 .1729712 0.28 0.777 .7580515 1.448587
v190 | .9311573 .0608004 -1.09 0.275 .8192111 1.058401
s708 | .9618182 .0489394 -0.77 0.444 .8704523 1.062774
s109 | .8670558 .1138723 -1.09 0.278 .6701333 1.121845
v513 | .6894548 .081396 -3.15 0.002 .5469255 .8691275
v024 | 1.043592 .0389911 1.14 0.254 .9698417 1.122952
v130 | 1.426153 .4571971 1.11 0.268 .7604202 2.674721
s801 | 1.833544 .7262324 1.53 0.126 .8430584 3.987722
v743d | 1.061066 .0604386 1.04 0.298 .9488905 1.186502
v731 | 1.015529 .0817099 0.19 0.848 .8672525 1.189156
------------------------------------------------------------------------------
. test v743=v731=0
Adjusted Wald test
v743 ambiguous abbreviation
r(111);
. test v743d=v731=0
Adjusted Wald test
( 1) v743d - v731 = 0
( 2) v743d = 0
F( 2, 1414) = 0.55
Prob > F = 0.5781
. test s801=v130=0
Adjusted Wald test
( 1) - v130 + s801 = 0
( 2) s801 = 0
F( 2, 1414) = 1.62
Prob > F = 0.1986
. test v024=v513=0
Adjusted Wald test
( 1) - v513 + v024 = 0
( 2) v024 = 0
F( 2, 1414) = 5.46
Prob > F = 0.0044
. test s109=s708=0
Adjusted Wald test
( 1) - s708 + s109 = 0
( 2) s109 = 0
F( 2, 1414) = 0.98
Prob > F = 0.3772
. test v013=v025=0
Adjusted Wald test
( 1) v013 - v025 = 0
( 2) v013 = 0
F( 2, 1414) = 0.11
Prob > F = 0.8954
. test v201=v106=0
Adjusted Wald test
( 1) - v106 + v201 = 0
( 2) v201 = 0
F( 2, 1414) = 1.93
Prob > F = 0.1459
.
*
* 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/
