Scilab Home page | Wiki | Bug tracker | Forge | Mailing list archives | ATOMS | File exchange
Scilab-Branch-6.1-GIT
Change language to: English - Français - Português - Русский

# hank

### 呼び出し手順

hk =hank(m, n, cov)

m

ブロック行の数

n

ブロック列の数

cov

hk

ハンケル行列の計算値

### 説明

この関数は,ベクトル過程の共分散系列から 大きさ(m*d,n*d)のハンケル行列を構築します. より正確には以下となります:

この関数は,ベクトル過程の共分散系列から 大きさ(m*d,n*d)のハンケル行列を作成します. より正しくは:

### 例

//Example of how to use the hank macro for

//building a Hankel matrix from multidimensional

//data (covariance or Markov parameters e.g.)

//

//This is used e.g. in the solution of normal equations

//by classical identification methods (Instrumental Variables e.g.)

//

//1)let's generate the multidimensional data under the form :

//  C=[c_0 c_1 c_2 .... c_n]

//where each bloc c_k is a d-dimensional matrix (e.g. the k-th correlation

//of a d-dimensional stochastic process X(t) [c_k = E(X(t) X'(t+k)], '

//being the transposition in scilab)

//

//we take here d=2 and n=64

c = rand(2, 2 * 64)

//generate the hankel matrix H (with 4 bloc-rows and 5 bloc-columns)

//from the data in c

H = hank(4, 5, c);

### 参照

• toeplitz — Toeplitz matrix (chosen constant diagonal bands)