corrarma(timeslab) | R Documentation |
Calculate ARMA Parameters from Autocorrelations
corrarma(rho,r0,p,q,maxit=100,del=1.e-5)
rho |
Array of length ${tt{p}}+{tt{q}}$ containing the autocorrelations of the process. |
r0 |
Real scalar containing the variance of the process $(>0)$. |
p |
Integer containing AR order $p(>0)$. |
q |
Integer containing MA order $q(>0)$. |
maxit |
Integer containing the maximum number of iterations to allow in Wilson's algorithm. |
del |
Real scalar containing convergence criterion $(>0)$. |
alpha |
Array of length {p } containing AR coefficients. |
beta |
Array of length {q } containing MA coefficients. |
rvar |
Real scalar containing the error variance $&sigma^2$. |
ier |
Integer variable containing an error/convergence indicator. The following values are possible: |
|
0 ~~ {CORRARMA } successfully found the ARMA parameters. |
|
1 ~~ A singular matrix was encountered trying to find AR parameters. |
|
2 ~~ Wilson's algorithm for finding the MA parameters didn't converge. |