1
lattn Scilab Group Scilab Function lattn
3
lattn - recursive solution of normal equations
8
n : maximum order of the filter
10
p : fixed dimension of the MA part. If p= -1, the algorithm reduces to
11
the classical Levinson recursions.
13
cov : matrix containing the Rk's (d*d matrices for a d-dimensional
14
process).It must be given the following way
22
la : list-type variable, giving the successively calculated polynomials
23
(degree 1 to degree n),with coefficients Ak
26
solves recursively on n (p being fixed) the following system (normal
27
equations), i.e. identifies the AR part (poles) of a vector ARMA(n,p)
30
|Rp+1 Rp+2 . . . . . Rp+n |
31
|Rp Rp+1 . . . . . Rp+n-1|
34
|I -A1 -A2 . . .-An|| . . . . . . . . |=0
37
| . . . . . . . Rp+1 |
38
|Rp+1-n. . . . . . Rp |
39
where {Rk;k=1,nlag} is the sequence of empirical covariances