use https://raw.github.com/remlapmot/mrrobust/master/dodata, clearHow to collect and export results
This example shows how to conveniently save and export your estimates using the r(table) matrix that is now returned by each command.
Setup example data
Select observations (p-value with exposure < 10-8)
gen byte sel1 = (ldlcp2 < 1e-8)Fit estimators - collecting results using r(table) matrix
IVW (with fixed effect standard errors)
mregger chdbeta ldlcbeta [aw=1/(chdse^2)] if sel1==1, ivw fe
mat ivw = r(table) Number of genotypes = 73
Residual standard error constrained at 1
------------------------------------------------------------------------------
| Coefficient Std. err. z P>|z| [95% conf. interval]
-------------+----------------------------------------------------------------
chdbeta |
ldlcbeta | .4815055 .038221 12.60 0.000 .4065938 .5564173
------------------------------------------------------------------------------MR-Egger (with SEs using an unconstrained residual variance)
mregger chdbeta ldlcbeta [aw=1/(chdse^2)] if sel1==1
mat mregger = r(table) Number of genotypes = 73
Residual standard error = 1.548
------------------------------------------------------------------------------
| Coefficient Std. err. z P>|z| [95% conf. interval]
-------------+----------------------------------------------------------------
chdbeta |
slope | .6173131 .1034573 5.97 0.000 .4145405 .8200858
_cons | -.0087706 .0054812 -1.60 0.110 -.0195136 .0019723
------------------------------------------------------------------------------MR-Egger using the radial formulation
mregger chdbeta ldlcbeta [aw=1/(chdse^2)] if sel1==1, radial
mat radial = r(table) Number of genotypes = 73
Residual standard error = 1.547
------------------------------------------------------------------------------
| Coefficient Std. err. z P>|z| [95% conf. interval]
-------------+----------------------------------------------------------------
radialGD |
radialGP | .642582 .1157871 5.55 0.000 .4156434 .8695205
_cons | -.5737301 .3545658 -1.62 0.106 -1.268666 .1212062
------------------------------------------------------------------------------Weighted mode estimator
mrmodal chdbeta chdse ldlcbeta ldlcse if sel1==1, weighted
mat mode = r(table) Number of genotypes = 73
Replications = 1000
Phi = 1
------------------------------------------------------------------------------
| Coefficient Std. err. z P>|z| [95% conf. interval]
-------------+----------------------------------------------------------------
beta | .4789702 .0663145 7.22 0.000 .3489963 .6089441
------------------------------------------------------------------------------Weighted median estimator
mrmedian chdbeta chdse ldlcbeta ldlcse if sel1==1, weighted
mat median = r(table) Number of genotypes = 73
Replications = 1000
------------------------------------------------------------------------------
| Coefficient Std. err. z P>|z| [95% conf. interval]
-------------+----------------------------------------------------------------
beta | .4582573 .0653722 7.01 0.000 .3301302 .5863845
------------------------------------------------------------------------------Combine and export results
Check our matrices
mat dir
mat list ivw
mat list mregger
mat list radial
mat list mode
mat list median median[9,1]
mode[9,1]
radial[9,2]
mregger[9,2]
ivw[9,1]
ivw[9,1]
chdbeta:
ldlcbeta
b .48150551
se .03822098
z 12.597937
pvalue 2.167e-36
ll .40659377
ul .55641726
df .
crit 1.959964
eform 0
mregger[9,2]
chdbeta: chdbeta:
slope _cons
b .61731315 -.00877065
se .10345735 .00548118
z 5.9668371 -1.60014
pvalue 2.419e-09 .10956752
ll .41454047 -.01951356
ul .82008582 .00197226
df . .
crit 1.959964 1.959964
eform 0 0
radial[9,2]
radialGD: radialGD:
radialGP _cons
b .64258196 -.57373006
se .11578709 .35456584
z 5.5496858 -1.61812
pvalue 2.862e-08 .10563675
ll .41564344 -1.2686663
ul .86952048 .12120621
df . .
crit 1.959964 1.959964
eform 0 0
mode[9,1]
beta
b .4789702
se .06631445
z 7.222712
pvalue 5.096e-13
ll .34899626
ul .60894415
df .
crit 1.959964
eform 0
median[9,1]
beta
b .45825733
se .0653722
z 7.0099725
pvalue 2.384e-12
ll .33013017
ul .58638448
df .
crit 1.959964
eform 0Combined into single matrix
mat output = (ivw, mregger, radial, mode, median)
mat colnames output = ivw_beta mregger_beta mregger_cons ///
radial_beta radial_cons mode_beta median_beta
mat coleq output = "" "" "" "" "" "" ""
mat output = output'
mat list output, format(%4.3f)output[7,9]
b se z pvalue ll ul df crit eform
ivw_beta 0.482 0.038 12.598 0.000 0.407 0.556 . 1.960 0.000
mregger_beta 0.617 0.103 5.967 0.000 0.415 0.820 . 1.960 0.000
mregger_cons -0.009 0.005 -1.600 0.110 -0.020 0.002 . 1.960 0.000
radial_beta 0.643 0.116 5.550 0.000 0.416 0.870 . 1.960 0.000
radial_cons -0.574 0.355 -1.618 0.106 -1.269 0.121 . 1.960 0.000
mode_beta 0.479 0.066 7.223 0.000 0.349 0.609 . 1.960 0.000
median_beta 0.458 0.065 7.010 0.000 0.330 0.586 . 1.960 0.000Export matrix to dataset
drop _all
svmat output, names(col)
local rownames : rownames output
di "`rownames'"
tokenize `rownames'
gen str15 estimate = ""
forvalues i = 1/7 {
replace estimate = "``i''" in `i'
}Show dataset
list estimate b se z pvalue ll ul, clean noobs estimate b se z pvalue ll ul
ivw_beta .4815055 .038221 12.59794 2.17e-36 .4065938 .5564172
mregger_beta .6173131 .1034573 5.966837 2.42e-09 .4145405 .8200858
mregger_cons -.0087707 .0054812 -1.60014 .1095675 -.0195136 .0019723
radial_beta .6425819 .1157871 5.549686 2.86e-08 .4156434 .8695205
radial_cons -.5737301 .3545658 -1.61812 .1056367 -1.268666 .1212062
mode_beta .4789702 .0663145 7.222712 5.10e-13 .3489963 .6089441
median_beta .4582573 .0653722 7.009973 2.38e-12 .3301302 .5863845 Save dataset
save myestimates, replaceExport as tab-delimited textfile
export delimited using myestimates.txt, delimiter(tab) replace