Hello Statalisters,
I'm having an issue obtaining the table that I want using collect with mi estimate and this issue only comes up with factor variables and not continuous ones.
I run the following :
output is shown below:
this is fine i can work with that table.
The issue arises when I am trying to use this in combination with mi estimate:
I notice that the _r_b series are not in the result dimensions with mi estimates using a factor variable but when I use a continuous variable with mi estimate they are indeed there and I get the table that I want...
I'd love to know what I'm doing wrong.
Regards,
Jean-Michel
I'm having an issue obtaining the table that I want using collect with mi estimate and this issue only comes up with factor variables and not continuous ones.
I run the following :
Code:
collect clear
collect TR=_r_b SE=_r_se Lower=_r_lb Upper=_r_ub Pvalue=_r_p, ///
:streg ib1.pbq_sex,vce(cluster team) dist(loglogistic) nolog tr
collect dims
collect levelsof result
collect style cell result, nformat(%6.3f)
collect layout (colname) (result)
Code:
. collect clear
. collect TR=_r_b SE=_r_se Lower=_r_lb Upper=_r_ub Pvalue=_r_p, ///
> :streg ib1.pbq_sex,vce(cluster team) dist(loglogistic) nolog tr
Failure _d: injuryr1==1 2 3 9
Analysis time _t: (enddate-origin)
Origin: time rtpdate
Exit on or before: time enddate
Loglogistic AFT regression
No. of subjects = 15,648 Number of obs = 15,648
No. of failures = 2,979
Time at risk = 1,213,389
Wald chi2(1) = 5.28
Log pseudolikelihood = -6890.3172 Prob > chi2 = 0.0215
(Std. err. adjusted for 313 clusters in team)
------------------------------------------------------------------------------
| Robust
_t | Time ratio std. err. z P>|z| [95% conf. interval]
-------------+----------------------------------------------------------------
pbq_sex |
Female | 1.77584 .4437184 2.30 0.022 1.088226 2.897933
_cons | 163.0145 11.89825 69.79 0.000 141.2856 188.0851
-------------+----------------------------------------------------------------
/lngamma | -.9076884 .1276073 -7.11 0.000 -1.157794 -.6575827
-------------+----------------------------------------------------------------
gamma | .4034558 .0514839 .3141784 .5181022
------------------------------------------------------------------------------
Note: Estimates are transformed only in the first equation to time ratios.
Note: _cons estimates baseline time.
. collect dims
Collection dimensions
Collection: default
-----------------------------------------
Dimension No. levels
-----------------------------------------
Layout, style, header, label
cmdset 1
coleq 3
colname 5
colname_remainder 1
pbq_sex 2
program_class 2
result 56
result_type 3
rowname 1
Style only
border_block 4
cell_type 4
-----------------------------------------
. collect levelsof result
Collection: default
Dimension: result
Levels: Lower N N_clust N_fail N_sub Pvalue SE TR Upper _r_b _r_ci _r_df _r_lb _r_p _r_se _r_ub _r_z
_r_z_abs chi2 chi2type clustvar cmd cmd2 cmdline converged dead depvar df_m frm2 gamma ic k
k_aux k_dv k_eq k_eq_model ll ll_0 ml_method opt p predict predict_sub properties rank rank0
rc risk stcurve t0 technique title user vce vcetype which
. collect style cell result, nformat(%6.3f)
. collect layout (colname) (result)
Collection: default
Rows: colname
Columns: result
Table 1: 4 x 5
-------------------------------------------------
| TR SE Lower Upper Pvalue
----------+--------------------------------------
Female | 1.776 0.444 1.088 2.898 0.022
Male | 1.000 0.000 . . .
lngamma | -0.908 0.128 -1.158 -0.658 0.000
Intercept | 163.014 11.898 141.286 188.085 0.000
-------------------------------------------------
.
this is fine i can work with that table.
The issue arises when I am trying to use this in combination with mi estimate:
Code:
. collect clear
. collect _r_b _r_se _r_lb _r_ub _r_p ///
> :mi estimate,dots eform(Time ratio) cmdok:streg ib1.pbq_sex,vce(cluster team) dist(loglogistic) n
> olog
Imputations (50):
.........10.........20.........30.........40.........50 done
Multiple-imputation estimates Imputations = 50
Loglogistic AFT regression Number of obs = 448
Average RVI = 0.0000
Largest FMI = 0.0000
DF adjustment: Large sample DF: min = 4.84e+61
avg = 4.84e+61
max = .
Model F test: Equal FMI F( 1, .) = 2.23
Within VCE type: Robust Prob > F = 0.1353
(Within VCE adjusted for 313 clusters in team)
------------------------------------------------------------------------------
_t | Time ratio Std. err. t P>|t| [95% conf. interval]
-------------+----------------------------------------------------------------
pbq_sex |
Female | 1.43824 .3499714 1.49 0.135 .8927003 2.317164
_cons | 174.1712 12.62802 71.17 0.000 151.0989 200.7666
-------------+----------------------------------------------------------------
/lngamma | -.8595957 .112937 -7.61 0.000 -1.080948 -.6382433
-------------+----------------------------------------------------------------
gamma | .4233332 .04781 .3392737 .5282196
------------------------------------------------------------------------------
Note: Estimates are transformed only in the first equation to time ratios.
Note: _cons estimates baseline time.
. collect dims
Collection dimensions
Collection: default
-----------------------------------------
Dimension No. levels
-----------------------------------------
Layout, style, header, label
cmdset 1
coleq 2
colname 9
colname_remainder 1
pbq_sex 2
program_class 1
result 99
result_type 3
roweq 2
rowname 5
Style only
border_block 4
cell_type 4
-----------------------------------------
. collect levelsof result
Collection: default
Dimension: result
Levels: B_mi Cns_mi F_mi M_mi N N_clust N_clust_mi N_fail N_max_mi N_mi N_min_mi N_sub N_sub_mi V_mi
W_mi _dfnote_mi _predict_mi _sortseed_mi _sortseedcmd_mi b_mi chi2 chi2type cilevel_mi
clustvar cmd cmd_mi cmdline cmdline_mi consonly converged converged_cons crittype dead depvar
df_avg_mi df_c_mi df_m df_m_mi df_max_mi df_mi df_min_mi df_r_mi dfadjust_mi diparm1 ecmd2_mi
ecmd_mi esampvary_mi fmi_max_mi fmi_mi frm2 gamma ic k k_autoCns k_aux k_dv k_eq k_eq_model
k_eq_model_mi k_exp_mi ll ll_0 m_est_mi m_mi marginsfootnote marginsprop mcerror_mi mi
ml_method modeltest_mi noconstant opt p p_mi pise_mi predict_sub prefix_mi rank rank0 rc
rc_mi re_mi reparm_rc_mi risk rvi_avg_F_mi rvi_avg_mi rvi_mi singularHmethod stcurve t0
technique title title_mi ufmi_mi user vce vcetype which wvce_mi
. collect style cell result, nformat(%6.3f)
. collect layout (colname) (result)
Collection: default
Rows: colname
Columns: result
Your layout specification does not identify any items.
I'd love to know what I'm doing wrong.
Regards,
Jean-Michel
Comment