ABSTRACT

FIGURE 4.3: Fitted average longitudinal proles for aortic gradient for the two types of operation for a patient with the median age.

The object Ps will in fact be the input data frame DF augmented with four extra columns with the tted values ^, their standard errors and the limits of the asymptotic 95% pointwise condence intervals. The plot of the resulting estimates is produced using function xyplot() from package lattice (Sarkar, 2008), and is shown in Figure 4.3.