IJSRP, Volume 4, Issue 10, October 2014 Edition [ISSN 2250-3153]
Viswanathan.N, Ravanan.R
Abstract:
Estimation of individual and population slopes using Linear Minimum Variance Unbiased Estimation by Wu and Bailey (1988) is applied for incomplete longitudinal studies. In these studies the right censoring process is considered informative, in which the number of observations made for each subject is assumed to vary depending on the rate of change or slope of the individual response variable. This process of ‘missing not at random’ poses problems in the estimation of the slope parameters. The proposed method provides better estimates under different parametric models for the censoring distribution. The method’s performance is studied through a simulation study under Standard Normal and Uniform censoring distributions.