EcoSta 2022: Start Registration
View Submission - EcoSta2022
A0433
Title: Optimal minimax rates of specification testing with data-driven bandwidth Authors:  Masamune Iwasawa - Otaru University of Commerce (Japan) [presenting]
Yoshihiko Nishiyama - Kyoto University (Japan)
Kohtaro Hitomi - Kyoto Institute of Technology (Japan)
Abstract: Optimal minimax rates of specification testing for linear and non-linear instrumental variable regression models are investigated. The test constructed by non-parametric kernel techniques can be rate optimal when bandwidths are selected appropriately. Since bandwidths are often selected in a data-dependent way in empirical studies, the rate optimality of the test with data-driven bandwidths are investigated. While the least-squares cross-validation selects bandwidths that are optimal for estimation, it is shown not to be optimal for testing. Thus, we propose a power maximizing bandwidth selection method that is optimal for testing.