Why the standadard errors of the coefficientes of the arma
models fited by using the arima procedure in the stats
package doesnt coincide with that of S+, minitab or SAS?
--
Este mensaje ha sido analizado por MailScanner
en busca de virus y otros contenidos peligrosos,
y se considera que
On Wed, 6 Dec 2006, Paulino Perez Rodriguez wrote:
Why the standadard errors of the coefficientes of the arma
models fited by using the arima procedure in the stats
package doesnt coincide with that of S+, minitab or SAS?
Possibly because R is using superior methodology (full ML fitting).
