Least squares cross-validation (CV) methods are often used for automated bandwidth selection. We show that they share a common structure which has an explicit asymptotic solution that we derive. Using the framework of density estimation, we consider unbiased, biased, and smoothed CV methods. We...

We review Bayesian inference for dynamic latent variable models using the data augmentation principle. We detail the difficulties of simulating dynamic latent variables in a Gibbs sampler. We propose an alternative specification of the dynamic disequilibrium model which leads to a simple...

We show how to perform Bayesian inference in dynamic disequilibrium models by data augmentation. Bayesian inference is much simpler than maximum likelihood estimation since multiple integrals that appear in the likelihood function are avoided by working in the spaces of latent and observed...