A1383
Title: Taming high dimensional cointegrated regressors
Authors: Ziwei Mei - University of Macau (China) [presenting]
Abstract: LASSO has been the state-of-the-art methodology for high dimensional linear regressions. When cointegration is present in a high dimensional predictive regression, LASSO is inconsistent due to the excessive penalty on cointegrated regressors. Nonconvex penalties, including the widely used SCAD and MCP, are proposed to fix the over-penalization and restore consistent estimation of the high dimensional predictive regressions. The consistency of estimation and variable selection by SCAD and MCP is established. In the numerical studies, substantial improvements by the nonconvex penalties relative to LASSO in terms of parameter estimation and out-of-sample prediction are observed.
