|PROC TPSPLINE with Large Data Sets|
The calculation of the penalized least squares estimate is computationally intensive. The amount of memory and CPU time needed for the analysis depends on the number of unique design points, which corresponds to the number of unknown parameters to be estimated.
PROC TPSPLINE selects one design point from the group and treats all observations in the group as replicates of that design point. Calculation of the thin-plate smoothing spline estimates is based on the reprocessed data. The way to choose the design point from a group depends on the order of the data. Hence, different orders of input data might result in different estimates.
By combining several design points into one, this option reduces the number of unique design points, thereby approximating the original data. The value you specify for the D= option determines the width of the range used to group the data.