Beyond curse of dimensionality with transition matrix

We now use some of the points on each grid and update:

k1=k[indM[indn,sind[t+1]]];

and

ks[t+1]=SmoothingFunction(k1,k[indn],ks[t]);

given k and k1 respectively starting from ks[startn]. Here, indn is an index vector

around ks[t] with nsmooth points.

We here set

0:linear interpolation

1:polynomial

2:LOESS

3:LOWESS

4:Kernel regression

5:Smoothing natural spline

6:NURBS x 10

7:Bezier Curve x 10