演講者簡介 : Professor Song received his Ph.D. in Economics from Universidad Carlos III de Madrid in 2014. He is currently an Assistant Professor in Guanghua School of Management, Peking University. His research fields are Econometric Theory, Nonparametric and Semiparametric Methods, Specification Testing, Bootstrap Methods, and Time Series Analysis.
演講摘要 : Neyman (1937)'s smooth test has proven to be an extremely valuable tool in the long history of statistical hypothesis testing. Smooth tests are inspired from the probability integral transform (PIT); for example, various smooth tests have been proposed to assess the goodness-of-fit of certain parametric distributions. Nevertheless, the majority of the existing literature focuses on PIT in parametric models, even although Neyman (1937)’s idea is easily applicable to PIT constructed from nonparametric models. In this talk I mainly discuss the promising aspects of the smooth tests for nonparametric models. In particular, I focus on smooth tests for (i) conditional independence, (ii) copula independence, and (iii) the equality of (conditional) distributions as well as the equality of copulas in the two-sample settings.
