<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<item>
  <id>01655682</id>
  <dt>j</dt>
  <an>01655682</an>
  <augroup>
    <au>Castro, Glaysar</au>
    <au>Girardin, Valerie</au>
  </augroup>
  <ti>Arma models realization and impulse responses.</ti>
  <so>Stoch. Models 17, No. 3, 293-312 (2001).</so>
  <py>2001</py>
  <pu>Taylor \& Francis, Philadelphia, PA</pu>
  <lagroup>
    <la>EN</la>
  </lagroup>
  <ccgroup>
  </ccgroup>
  <utgroup>
    <ut>auto-regressive moving average models</ut>
    <ut>covariances</ut>
    <ut>impulse response</ut>
    <ut>maximum entropy</ut>
    <ut>order identification</ut>
    <ut>spectral density</ut>
    <ut>stationary discrete-time process</ut>
  </utgroup>
  <cigroup>
  </cigroup>
  <ligroup>
    <li>doi:10.1081/STM-100002275</li>
  </ligroup>
  <abgroup>
    <ab>Summary: A parallel is made between the role played by covariances in the determination of auto-regressive models and the role played by impulse responses in the determination of ARMA models. Autoregressive models are known to maximize the Burg-entropy under covariance constraints. Autoregressive-moving-average models give the maximum of the Burg-entropy among processes sharing the same covariances and impulse responses up to a certain lag. Such models are constructed by iterative or algebraic methods under different constraints. A new recursive method of identification of the order of an ARMA model is also developed, based on the generalized reflection coefficients</ab>
    <rv></rv>
  </abgroup>
</item>