Goodness-of-Fit for Conditional Distributions: An Approach Using Principal Component Analysis and Component Selection

Description

The paper presents a novel goodness-of-fit test technique for parametric conditional distributions. The authors develop a method based on a residual marked empirical process, employing conditional Principal Component Analysis (PCA) to extract components that serve as a basis for various types of new tests, including component tests and smooth tests. The paper also introduces a component selection approach to identify the most contributory components, thereby improving testing power.

Period	29 Jun 2024
Held at	Asian Meeting of the Econometric Society
Degree of Recognition	International