1
.TH armax1 1 "April 1993" "Scilab Group" "Scilab Function"
4
armax1 - armax identification
7
[a,b,d,sig,resid]=armax1(r,s,q,y,u,[b0f])
18
: auto regression orders with r >=0, s >=-1.
21
: optional parameter. Its default value is 0 and it means that the coefficient b0 must be identified. if bof=1 the b0 is supposed to be zero and is not identified
24
: is the vector [1,a1,...,a_r]
27
: is the vector [b0,......,b_s]
30
: is the vector [1,d1,....,d_q]
33
: resid=[ sig*echap(1),....,];
35
armax1 is used to identify the coefficients of a 1-dimensional
38
A(z^-1)y= B(z^-1)u + D(z^-1)sig*e(t)
39
e(t) is a 1-dimensional white noise with variance 1.
40
A(z)= 1+a1*z+...+a_r*z^r; ( r=0 => A(z)=1)
41
B(z)= b0+b1*z+...+b_s z^s ( s=-1 => B(z)=0)
42
D(z)= 1+d1*z+...+d_q*z^q ( q=0 => D(z)=1)
44
for the method, see Eykhoff in trends and progress in system identification) page 96.
46
z(t)=[y(t-1),..,y(t-r),u(t),...,
47
u(t-s),e(t-1),...,e(t-q)]
49
coef= [-a1,..,-ar,b0,...,b_s,d1,...,d_q]'
50
y(t)= coef'* z(t) + sig*e(t).
52
a sequential version of the AR estimation where e(t-i) is replaced
53
by an estimated value is used (RLLS). With q=0 this method is exactly
54
a sequential version of armax