A discussion on: Approximate Bayesian inference for latent Gaussian models by using integrated nested Laplace approximations by Rue, H. Martino, S. and Chopin, N.

This is a discussion of the work Rue et al. (2009). In order to evaluate the impact of their Gaussian approximation on the marginal posterior we consider here a slightly different albeit standard stochastic volatility model. We propose a pluggin approximation that is readily available, contrary to the mode of the full conditional suggested in Rue et al. (2009). We obtain a straightforward recurrence relations.