A0507
Title: The Dantzig selector for semiparametric models of stochastic processes
Authors: Kou Fujimori - Shinshu University (Japan) [presenting]
Koji Tsukuda - Kyushu University (Japan)
Abstract: The sparse estimation problem is considered for models of stochastic processes with possibly infinite-dimensional nuisance parameters by using the Dantzig selector, which can be seen as an extension of the Z-estimator. When a consistent estimator for a nuisance parameter is obtained, it is possible to construct an asymptotically normal estimator for the parameter of interest under appropriate conditions. Motivated by this fact, the asymptotic behaviour of the Dantzig selector is established for models of ergodic stochastic processes with high-dimensional parameters of interest and possibly infinite-dimensional nuisance parameters. The applications for ergodic diffusion processes and ergodic time series, including integer-valued autoregressive models, are presented.
