Parameter inference on autoregressive models is usually carried out conditi
onally on a previously selected lag order. In the majority of cases the lag
order selection is carried out using information criteria and in particula
r the Akaike [2nd International Symposium on Information Theory (1973) 267-
281], Schwarz [Annuls of Statistics (1978) 461-464] or Hannan and Quinn [Jo
urnal of the Royal Statistical Society (Series B), 41 (1979) 190-195] crite
ria. It is well known that the latter two criteria are consistent in lag or
der selection in the sense of picking the true order of the system with pro
bability one asymptotically. On the other hand, Akaike's criterion is known
to overestimate the lag order in this sense. In this note we discuss the a
symptotic distribution, of the parameter estimates without conditioning on
the lag order selected. (C) 2001 Elsevier Science BN. All rights reserved.