 |
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
st: -bs- with the xi: prefix and -gologit2-
Please forgive this basic question. I'm not very confident with
bootstrapping.
I am trying to get an estimate of bias in the regression coefficients
from a partial proportional odds model that I have fitted using
-gologit2-. A colleague tells me that she has previously obtained an
estimate of bias from the -logit- function using -bs- in Stata 7. I can
replicate this (with -logit-) as long as I don't have any indicator
variables (i.e. I don't use the xi: prefix) in Stata 9.2. However, the
output I get when I use -bs- with -gologit2- doesn't make much sense, I
get too many sets of coefficients (5, when I only 3 levels to my
dependent variable - see output below).
. bs "gologit2 cohortsev agegrp1 agegrp2 agegrp3 gender ass_4 rad_k
ffdef knflex, npl(gender)" _b _se
command: gologit2 cohortsev agegrp1 agegrp2 agegrp3 gender ass_4
rad_k ffdef knflex , npl(gender)
statistics: b_agegrp2 = [None]_b[agegrp2]
b_agegrp3 = [None]_b[agegrp3]
b_gender = [None]_b[gender]
b_ass_4 = [None]_b[ass_4]
b_rad_k = [None]_b[rad_k]
b_ffdef = [None]_b[ffdef]
b_knflex = [None]_b[knflex]
b_cons = [None]_b[_cons]
b_1agegrp2 = [Mild]_b[agegrp2]
b_1agegrp3 = [Mild]_b[agegrp3]
b_1gender = [Mild]_b[gender]
b_1ass_4 = [Mild]_b[ass_4]
b_1rad_k = [Mild]_b[rad_k]
b_1ffdef = [Mild]_b[ffdef]
b_1knflex = [Mild]_b[knflex]
b_1cons = [Mild]_b[_cons]
se_agegrp2 = [None]_se[agegrp2]
se_agegrp3 = [None]_se[agegrp3]
se_gender = [None]_se[gender]
se_ass_4 = [None]_se[ass_4]
se_rad_k = [None]_se[rad_k]
se_ffdef = [None]_se[ffdef]
se_knflex = [None]_se[knflex]
se_cons = [None]_se[_cons]
se_1ageg~2 = [Mild]_se[agegrp2]
se_1ageg~3 = [Mild]_se[agegrp3]
se_1gender = [Mild]_se[gender]
se_1ass_4 = [Mild]_se[ass_4]
se_1rad_k = [Mild]_se[rad_k]
se_1ffdef = [Mild]_se[ffdef]
se_1knflex = [Mild]_se[knflex]
se_1cons = [Mild]_se[_cons]
note: label truncated to 80 characters
Bootstrap statistics Number of obs
= 695
Replications
= 50
------------------------------------------------------------------------------
Variable | Reps Observed Bias Std. Err. [95% Conf. Interval]
-------------+----------------------------------------------------------------
b_agegrp2 | 50 .8023604 -.8493435 .8232245 -.8519712
2.456692 (N)
| -1.029455
1.120293 (P)
| -.7181095
1.149808 (BC)
b_agegrp3 | 23 1.613888 .0510502 .219334 1.159017
2.068759 (N)
| 1.362029
2.222399 (P)
| 1.362029
2.222399 (BC)
b_gender | 50 .7143635 .0192722 .2133874 .2855455
1.143182 (N)
| .3096491
1.162701 (P)
| .2976162
1.162701 (BC)
b_ass_4 | 50 .410215 -.0474633 .2559171 -.1040697
.9244998 (N)
| -.0704969
.8251594 (P)
| -.0490323
.9484665 (BC)
b_rad_k | 50 -.6291185 -.0027164 .2513221 -1.134169
-.1240678 (N)
| -1.166821
-.1492876 (P)
| -1.166821
-.1492876 (BC)
b_ffdef | 50 1.477184 .0457175 .3246976 .8246799
2.129688 (N)
| .9675916
2.039259 (P)
| .6020367
1.888914 (BC)
b_knflex | 50 1.39712 -.0603133 .2741078 .8462799
1.947961 (N)
| .8755013
1.961069 (P)
| .9153236
2.106204 (BC)
b_cons | 50 -3.772725 .9204005 1.068858 -5.920676
-1.624773 (N)
| -4.847654
-.9911599 (P)
| -4.865833
-2.332712 (BC)
b_1agegrp2 | 50 .8023604 -.8493435 .8232245 -.8519712
2.456692 (N)
| -1.029455
1.120293 (P)
| -.7181095
1.149808 (BC)
b_1agegrp3 | 23 1.613888 .0510502 .219334 1.159017
2.068759 (N)
| 1.362029
2.222399 (P)
| 1.362029
2.222399 (BC)
b_1gender | 50 -.0356487 .0117261 .1829494 -.4032992
.3320018 (N)
| -.3514609
.2623091 (P)
| -.4675419
.2623091 (BC)
b_1ass_4 | 50 .410215 -.0474633 .2559171 -.1040697
.9244998 (N)
| -.0704969
.8251594 (P)
| -.0490323
.9484665 (BC)
b_1rad_k | 50 -.6291185 -.0027164 .2513221 -1.134169
-.1240678 (N)
| -1.166821
-.1492876 (P)
| -1.166821
-.1492876 (BC)
b_1ffdef | 50 1.477184 .0457175 .3246976 .8246799
2.129688 (N)
| .9675916
2.039259 (P)
| .6020367
1.888914 (BC)
b_1knflex | 50 1.39712 -.0603133 .2741078 .8462799
1.947961 (N)
| .8755013
1.961069 (P)
| .9153236
2.106204 (BC)
b_1cons | 50 -4.206401 .9271253 1.043046 -6.302479
-2.110322 (N)
| -5.076808
-1.548373 (P)
| -5.128612
-2.800457 (BC)
se_agegrp2 | 50 .1771546 .0077713 .0072581 .162569
.1917403 (N)
| .173128
.1969959 (P)
| .1710746
.1880535 (BC)
se_agegrp3 | 23 .2038915 .002957 .0034246 .1967893
.2109936 (N)
| .2009553
.2119513 (P)
| .2009553
.2098833 (BC)
se_gender | 50 .1850463 .0015962 .004902 .1751954
.1948973 (N)
| .1800057
.1978476 (P)
| .1794556
.1946131 (BC)
se_ass_4 | 50 .2140505 .001657 .0116807 .1905771
.2375238 (N)
| .1998482
.2409628 (P)
| .1998482
.2409628 (BC)
se_rad_k | 50 .2326257 .0043931 .0115427 .2094297
.2558217 (N)
| .2128248
.2567449 (P)
| .2114028
.2480152 (BC)
se_ffdef | 50 .2865268 .003983 .0248618 .2365652
.3364884 (N)
| .2530232
.3446471 (P)
| .2467398
.3446471 (BC)
se_knflex | 50 .259986 .0014732 .0166322 .2265624
.2934096 (N)
| .2325539
.2922703 (P)
| .2314252
.2922703 (BC)
se_cons | 50 .5996358 .0094903 .0215319 .5563658
.6429057 (N)
| .5705682
.6594479 (P)
| .5588557
.6280385 (BC)
se_1agegrp2 | 50 .1771546 .0077713 .0072581 .162569
.1917403 (N)
| .173128
.1969959 (P)
| .1710746
.1880535 (BC)
se_1agegrp3 | 23 .2038915 .002957 .0034246 .1967893
.2109936 (N)
| .2009553
.2119513 (P)
| .2009553
.2098833 (BC)
se_1gender | 50 .1761007 .0017658 .003134 .1698026
.1823987 (N)
| .1727841
.1837342 (P)
| .1725901
.1803827 (BC)
se_1ass_4 | 50 .2140505 .001657 .0116807 .1905771
.2375238 (N)
| .1998482
.2409628 (P)
| .1998482
.2409628 (BC)
se_1rad_k | 50 .2326257 .0043931 .0115427 .2094297
.2558217 (N)
| .2128248
.2567449 (P)
| .2114028
.2480152 (BC)
se_1ffdef | 50 .2865268 .003983 .0248618 .2365652
.3364884 (N)
| .2530232
.3446471 (P)
| .2467398
.3446471 (BC)
se_1knflex | 50 .259986 .0014732 .0166322 .2265624
.2934096 (N)
| .2325539
.2922703 (P)
| .2314252
.2922703 (BC)
se_1cons | 50 .6011292 .0097523 .0215747 .5577733
.6444852 (N)
| .5780582
.6642864 (P)
| .5639434
.6284492 (BC)
------------------------------------------------------------------------------
I can use -xi:- with -bootstrap- and -gologit2-, but this doesn't give
me an estimate of bias. I thought there might be an option to specify to
ask for an estimate of bias using this command, but I can't work out
what it is. Does anyone know what this might be please, or of another
way I can get this value?
. xi: bootstrap, reps(5): gologit2 cohortsev i.agegrp gender i.bmigrp ass_4 rad_k i.k8 i.eff ffdef knflex i.ccrep, npl(gend
er _Iage*)
i.agegrp _Iagegrp_1-3 (naturally coded; _Iagegrp_1 omitted)
i.bmigrp _Ibmigrp_1-3 (naturally coded; _Ibmigrp_1 omitted)
i.k8 _Ik8_1-4 (naturally coded; _Ik8_1 omitted)
i.eff _Ieff_1-3 (naturally coded; _Ieff_1 omitted)
i.ccrep _Iccrep_1-3 (naturally coded; _Iccrep_1 omitted)
(running gologit2 on estimation sample)
Bootstrap replications (5)
----+--- 1 ---+--- 2 ---+--- 3 ---+--- 4 ---+--- 5
.....
Generalized Ordered Logit Estimates Number of obs = 695
Replications = 5
Wald chi2(4) = 56.68
Prob > chi2 = 0.0000
Log likelihood = -590.14037 Pseudo R2 = 0.2204
( 1) [None]_Ibmigrp_2 - [Mild]_Ibmigrp_2 = 0
( 2) [None]_Ibmigrp_3 - [Mild]_Ibmigrp_3 = 0
( 3) [None]ass_4 - [Mild]ass_4 = 0
( 4) [None]rad_k - [Mild]rad_k = 0
( 5) [None]_Ik8_2 - [Mild]_Ik8_2 = 0
( 6) [None]_Ik8_3 - [Mild]_Ik8_3 = 0
( 7) [None]_Ik8_4 - [Mild]_Ik8_4 = 0
( 8) [None]_Ieff_2 - [Mild]_Ieff_2 = 0
( 9) [None]_Ieff_3 - [Mild]_Ieff_3 = 0
(10) [None]ffdef - [Mild]ffdef = 0
(11) [None]knflex - [Mild]knflex = 0
(12) [None]_Iccrep_2 - [Mild]_Iccrep_2 = 0
(13) [None]_Iccrep_3 - [Mild]_Iccrep_3 = 0
------------------------------------------------------------------------------
| Observed Bootstrap Normal-based
cohortsev | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
None |
_Iagegrp_2 | .7123798 .2264505 3.15 0.002 .2685449 1.156215
_Iagegrp_3 | 1.043262 .3293716 3.17 0.002 .3977058 1.688819
gender | .7721331 .2206287 3.50 0.000 .3397087 1.204557
_Ibmigrp_2 | .591031 .220149 2.68 0.007 .1595469 1.022515
_Ibmigrp_3 | 1.178145 .0788703 14.94 0.000 1.023562 1.332728
ass_4 | .4632416 .1068037 4.34 0.000 .2539102 .672573
rad_k | -.8335499 .1778562 -4.69 0.000 -1.182142 -.4849582
_Ik8_2 | .2250947 .1985405 1.13 0.257 -.1640375 .6142268
_Ik8_3 | 1.034946 .1822769 5.68 0.000 .6776894 1.392202
_Ik8_4 | .9789583 .3139054 3.12 0.002 .3637149 1.594202
_Ieff_2 | 1.190831 .2221256 5.36 0.000 .7554726 1.626189
_Ieff_3 | .7200264 .3638055 1.98 0.048 .0069808 1.433072
ffdef | 1.351051 .3402529 3.97 0.000 .6841678 2.017935
knflex | .884254 .2231974 3.96 0.000 .4467951 1.321713
_Iccrep_2 | .3234199 .2591339 1.25 0.212 -.1844732 .831313
_Iccrep_3 | 1.046906 .2365906 4.42 0.000 .5831972 1.510615
_cons | -4.456467 .6726127 -6.63 0.000 -5.774763 -3.13817
-------------+----------------------------------------------------------------
Mild |
_Iagegrp_2 | .7309714 .2866294 2.55 0.011 .169188 1.292755
_Iagegrp_3 | 1.809441 .3000004 6.03 0.000 1.221451 2.397431
gender | -.0449617 .2200139 -0.20 0.838 -.4761811 .3862577
_Ibmigrp_2 | .591031 .220149 2.68 0.007 .1595469 1.022515
_Ibmigrp_3 | 1.178145 .0788703 14.94 0.000 1.023562 1.332728
ass_4 | .4632416 .1068037 4.34 0.000 .2539102 .672573
rad_k | -.8335499 .1778562 -4.69 0.000 -1.182142 -.4849582
_Ik8_2 | .2250947 .1985405 1.13 0.257 -.1640375 .6142268
_Ik8_3 | 1.034946 .1822769 5.68 0.000 .6776894 1.392202
_Ik8_4 | .9789583 .3139054 3.12 0.002 .3637149 1.594202
_Ieff_2 | 1.190831 .2221256 5.36 0.000 .7554726 1.626189
_Ieff_3 | .7200264 .3638055 1.98 0.048 .0069808 1.433072
ffdef | 1.351051 .3402529 3.97 0.000 .6841678 2.017935
knflex | .884254 .2231974 3.96 0.000 .4467951 1.321713
_Iccrep_2 | .3234199 .2591339 1.25 0.212 -.1844732 .831313
_Iccrep_3 | 1.046906 .2365906 4.42 0.000 .5831972 1.510615
_cons | -5.266949 .7475724 -7.05 0.000 -6.732164 -3.801734
------------------------------------------------------------------------------
Best wishes,
Sara
Sara Mottram
Research Assistant: Biostatistics
Primary Care Musculoskeletal Research Centre
Primary Care Sciences
Keele University
Staffordshire, ST5 5BG
Tel: +44 (0) 1782 584711
Fax: +44 (0) 1782 583911
Email: [email protected]
*
* 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/