Performance of weighted estimating equations for longitudinal binary data with drop-outs missing at random.

Stat Med 2002 Oct;21(20):3035-54

Department of Biostatistics, CB #7420, School of Public Health, University of North Carolina, Chapel Hill 27599, USA.

The generalized estimating equations (GEE) approach is commonly used to model incomplete longitudinal binary data. When drop-outs are missing at random through dependence on observed responses (MAR), GEE may give biased parameter estimates in the model for the marginal means. A weighted estimating equations approach gives consistent estimation under MAR when the drop-out mechanism is correctly specified. In this approach, observations or person-visits are weighted inversely proportional to their probability of being observed. Using a simulation study, we compare the performance of unweighted and weighted GEE in models for time-specific means of a repeated binary response with MAR drop-outs. Weighted GEE resulted in smaller finite sample bias than GEE. However, when the drop-out model was misspecified, weighted GEE sometimes performed worse than GEE. Weighted GEE with observation-level weights gave more efficient estimates than a weighted GEE procedure with cluster-level weights.

Download full-text PDF

Source
http://dx.doi.org/10.1002/sim.1241DOI Listing
October 2002
54 Reads

Publication Analysis

Top Keywords

weighted gee
20
estimating equations
12
gee
9
binary data
8
missing random
8
longitudinal binary
8
data drop-outs
8
drop-outs missing
8
weighted estimating
8
weighted
7
performance unweighted
4
study compare
4
compare performance
4
unweighted weighted
4
mar drop-out
4
repeated binary
4
binary response
4
time-specific repeated
4
models time-specific
4
gee models
4

References

(Supplied by CrossRef)

Diggle et al.
1994
Article in Biometrika
Liang et al.
Biometrika 1986
Article in Statistics in Medicine
Hendricks et al.
Statistics in Medicine 1996
Article in Biometrics
Carr et al.
Biometrics 1993
Article in Biometrics
Bieler et al.
Biometrics 1995
Article in Journal of Statistical Computation and Simulation
Sharples et al.
Journal of Statistical Computation and Simulation 1992
Article in Journal of Statistical Computation and Simulation
Emrich et al.
Journal of Statistical Computation and Simulation 1992
Article in Biometrics
Lipsitz et al.
Biometrics 1994
Article in Communications in Statistics-Simulations
Gunsolley et al.
Communications in Statistics-Simulations 1995
Article in Biometrics
Lipsitz et al.
Biometrics 2000
Article in Journal of the American Statistical Association
Preisser et al.
Journal of the American Statistical Association 2000

Similar Publications