Growth curve analysis of Effect of Maternal Cocaine Use on Child Growth
Recently a lot of attention has been given to the analysis of longitudinal data
and repeated measures. Several issues dealing with il-conditioned
observations, ie, missing observations, censored observations and
irregular sampling have been addressed in the recent literature. We
revisit the condition where observations are made at irregular
sampling time and offer a comparable and intrinsic solution to this problem.
To show this procedure we will describe the analysis of growth curves generated from a cohort of children
measured from birth using three characteristics: height (centimeters), weight (in kilograms), and head circumference (centimeters). The
significant chracteristic being that of an irregular sampling rate.
We address these issues in assessing the association of maternal cocaine abuse
on the observations.