Usage: |
{A,V,ctrl}=iteras(A0,S1,S2,b,N,eps,itmax,expl)
|
Input: |
| A0 | p x p matrix, starting value for A;
if 0 is entered I^{-1} is taken
if p>1 and dim(A0)==1, I^{-1}*A0 is taken
|
| S1 | p x p matrix, Covariance of the first (clipped) component
|
| S2 | p x p matrix, Covariance of the second (unclipped) component
|
| b | numeric, clipping height
|
| N | scalar, MC-sample size / integration grid-points
|
| eps | numeric, exactitude
|
| itmax | scalar, maximal number of iterations
|
| expl | numeric; threshold for the changes in absolute
value of A: beyond this value convergence is uncertain ~ 4
|
Output: |
| A | p x p matrix, Lagrange-Multiplier solving (1) |
| V | p x p matrix, corresponding Covariance |
| ctrl | scalar, tells if convergence "happened" |