Francesca Rossi on Indirect Inference in Spatial Autoregression

Relatore:  Francesca Rossi - University of Southampton
  lunedì 15 dicembre 2014 alle ore 12.30 Aula Menegazzi Palazzo Economia
Ordinary least squares (OLS) is well-known to produce an inconsistent
estimator of the spatial parameter in pure spatial autoregression (SAR).
This paper explores the potential of indirect inference to correct the in-
consistency of OLS. Under broad conditions, it is shown that indirect
inference (II) based on OLS produces consistent and asymptotically nor-
mal estimates in pure SAR regression. The II estimator is robust to
departures from normal disturbances and is computationally straightfor-
ward compared with pseudo Gaussian maximum likelihood (PML). Monte
Carlo experiments based on various speci cations of the weighting matrix
con rm that the indirect inference estimator displays little bias even in
very small samples and gives overall performance that is comparable to
the Gaussian PML.
Documenti
Titolo Formato  (Lingua, Dimensione, Data pubblicazione)
Paper  pdfpdf (en, 621 KB, 18/11/14)

Referente
Francesco De Sinopoli

Referente esterno
Marcella Veronesi

Data pubblicazione
24 settembre 2014