拉斯维加斯9888

Title（标题）：Specification Testing for Nonparametric Structural Models with Monotonicity in Unobservables

Speaker（汇报人）： Professor Liangjun Su, School of Economics, Singapore Management University

Time（功夫）：2011年5月19日（周四）下午2:00 — 3:00

Place（地址）：拉斯维加斯9888新楼217教室

Abstract（提要）：

Monotonicity in a scalar unobservable is a crucial identifying assumption for an important class of nonparametric structural models accommodating unobserved heterogeneity, as in, for example, Altonji and Matzkin (2005) and Imbens and Newey (2009). Tests for this monotonicity have previously been unavailable. In this paper we propose and analyze tests for scalar monotonicity in nonparametric structural models under the assumption of conditional exogeneity. As it turns out, our tests have well behaved limiting distributions under the null, are consistent against precisely specified alternatives, and have standard local power properties. We provide straightforward bootstrap methods for inference. Some Monte Carlo experiments show that, for empirically relevant sample sizes, these reasonably control the level of the test, and that our tests have useful power. We apply our tests to study the monotonicity of ability in explaining the black-white earnings gap and that of unobserved preference in explaining the Engel curves.