Estimation of Count Data Models with Multiplicative Fixed Effects and Spatial Dependence

This note looks at estimation of count models with multiplicative fixed effects and spatial dependence in the form of spatial lag autocorrelation. We show that in presence of significant proportion of zeros in the count variable, control function estimation to model such spatial dependence is biased and inconsistent and requires change to the `assumed' form of proxy for spatial autoregressive lag. We also present a Conditional Poisson Maximum Likelihood approach based on `reduced' form conditional mean that yields consistent estimates