In the study of longitudinal twin and family data, interest is often in the
covariance structure of the data and the decomposition of this covariance
structure into genetic and environmental components rather than in estimati
ng the mean function. Various parametric models for covariance structures h
ave been proposed but, e.g., in studies of children where growth spurts occ
ur at various ages, it is difficult to a priori determine an appropriate pa
rametric model for the covariance structure. In particular, there is a gene
ral lack of the visualization procedures, such as lowess, that are invaluab
le in the initial stages of constructing a parametric model for a mean func
tion. Here we use kernel smoothing to modify a cross-sectional approach bas
ed on the sample covariance matrices to obtain smoothed estimates of the ge
netic and environmental variances and correlations for longitudinal twin da
ta. The methods are proposed to be exploratory as an aid to parametric mode
ling rather than inferential, although approximate asymptotic standard erro
rs are derived in the Appendix.