Counterexamples to 'On estimating noncausal nonminimum phase ARMA models of non-Gaussian processes' (and reply)

Abstract

In the above-named work by Giannakis and Swami (see ibid., vol.38, no.3, p.478, 1990), an order selection procedure was proposed for parameter estimation for noncausal, nonminimum phase ARMA (autoregressive moving average) models of non-Gaussian processes. The commenter argues that it has been derived under an erroneous assumption, and he also gives a counterexample to show that it does not yield a consistent order estimate in general. Two linear approaches for parameter estimation were also presented in the above-named work. The commenter contends that an existing counterexample to an earlier version of one of the algorithms also applies to both the approaches of the above-named work. The authors' reply is presented. >