help metandi postestimation(Roger Harbord) also see: metandi -------------------------------------------------------------------------------

Title

metandi postestimation-- postestimation tools for metandi

DescriptionMany of Stata's standard postestimation tools will not work after

metandi, or won't work as expected, asmetandireshapes the data before fitting the model.The notable exception is

predict, which can be used to obtain posterior predictions (empirical Bayes estimates) of the sensitivity and specificity in each study (mu), as well as various statistics that may be useful for detecting outliers (e.g.ustd) and influential observations (cooksd).

metandiplotgraphs the results frommetandi.

Syntax for predict

predict[type]newvarlist[if] [in] [,statistic]where

statisticis

muposterior predicted (empirical Bayes) sensitivity and specificity; the defaultuposterior means (empirical Bayes predictions, BLUPs) of random effectssduposterior standard deviations of the random effectsustdstandardized posterior means of random effectslinpredlinear predictor with empirical Bayes predictions plugged in. linpred = xb + ucooksdCook's distance for each study. Only available when model was fitted usinggllamm.

Most of the above statistics require two new variables to store them, one for the statistic associated with sensitivity, and one for the statistic associated with specificity. If

newvarlistcontains only onenewvar, the statistics associated with sensitivity and specificity will be stored innewvar1andnewvar0respectively.cooksd, however, is computed once for each study and therefore requires only onenewvar.

RemarksIf the model was fitted using

gllamm, the predictions are obtained usinggllapred, whose help page contains further details of the statistics available. See also the GLLAMM manual (Rabe-Hesketh, Skrondal & Pickles 2004) and Rabe-Hesketh & Skrondal (2005). If the model was fitted usingxtmelogit, the predictions are obtained usingpredict- see xtmelogit postestimation##predict.

Examples

. metandi tp fp fn tnAdd empirical Bayes estimates to

metandiplot:

. predict eb

. metandiplot, addplot(scatter eb1 eb0)Check for particularly influential observations using Cook's distance:

. predict cook, cooksd

. scatter cook studyid, mlabel(author)Check for outliers using standardized predicted random effects (interpretable as standardized level-2 residuals):

. predict ustd_Se ustd_Sp, ustd

. scatter ustd_Se ustd_Sp, xscale(rev) mlab(studyid) xline(0)yline(0)

Also seeOnline:

gllapred,metandi,metandiplot