1. What are the future works in "A generalized moments estimator for the autoregressive parameter in a spatial model" ?
It should be of interest to extend the generalized moments approach suggested in this paper to those models, and to determine corresponding large sample properties.
read more
2. What have the authors contributed in "A generalized moments estimator for the autoregressive parameter in a spatial model" ?
This paper is concerned with the estimation of the autoregressive parameter in a widely considered spatial autocorrelation model.. However, as discussed in the paper, the ( quasi ) maximum likelihood estimator may not be computationally feasible in many cases involving moderate or large sized samples.. In this paper the authors suggest a generalized moments estimator that is computationally simple irrespective of the sample size.. The authors provide results concerning the large and small sample properties of this estimator.
read more
3. What is the log-likelihood for the model in (1)?
Assuming for the moment that u is observable and normally distributed, the log-likelihood for the model in (1) is, using evident notation, given byln(L) = − N2 {ln(σ2) + ln(2π)} (3)− 12σ2 u′(I − ρM ′)(I − ρM)u+ ln ||I − ρM || .
read more
4. What is the significance of the use of idealized weighting matrices?
Of course, the use of idealized weighting matrices raises the concern that results corresponding to those matrices may15The authors note that ρ̂QML (and σ̂ 2 QML) denote the (joint) maximizers of the normal loglikelihood function (3), even if the actual distribution is not normal.
read more
