Estimation of Spatial Autoregressive Models for Origin-Destination Flows : A Partial Likelihood Approach

We extend LeSage and Pace's (2008) spatial autoregressive model for origin-destination flows by accommodating two-way fixed effects. Those fixed effects represent unobserved characteristics of origin and destination units. A partial likelihood approach is used to remove fixed effects in the model. The quasi-maximum likelihood estimator of the partial likelihood function is consistent and asymptotically centered normal. Monte Carlo experiments verify this advantage in finite samples. Using our specification, we capture significant spatial influences, but their magnitudes are smaller than those from the model without fixed effects