Title: Sparse indirect inference
Authors: Paola Stolfi - CNR - Institute for Applied Mathematics (Italy) [presenting]
Mauro Bernardi - University of Padova (Italy)
Lea Petrella - Sapienza University of Rome (Italy)
Abstract: Indirect inference methods are simulation-based procedures for estimating the parameters of an intractable model using an alternative auxiliary model which is computationally simpler. Such methods have received a lot of attention in literature, but no attempt has been made to deal with large dimensional problems. We introduce the sparse indirect inference estimator by adding the Smoothly Clipped Absolute Deviation penalty. We establish consistency and asymptotic normality of the proposed estimator and we show that the Sparse-Indirect Inference estimator enjoys the oracle properties under mild regularity conditions. The method is applied to estimate the parameters of large dimensional Non-Gaussian dynamic regression models.