How powerful is the antedependence test for analysing serial measurements with dropouts?

In this article suitability of the antedependence test for analysing serial measurements in the presence of dropouts is studied. First, as an example, the dropout process for postoperative pain measurements is modelled. Then, using simulation techniques, power of the antedependence test is assessed in the presence of different types of dropout processes and different percentages of dropouts. Finally, the power of antedependence test in the presence of dropouts is exemplified by a real data-set. The results show that the power of the antedependence test is satisfactorily high even with as much as 50 percent of dropouts. © 2008 Nova Science Publishers, Inc.