期刊名称：CEMFI Working Papers / Centro de Estudios Monetarios y Financieros, Madrid
出版年度：2017
卷号：2017
出版社：Centro de Estudios Monetarios y Financieros, Madrid
摘要：We exploit the rationale behind the Expectation Maximization algorithm to derive simple to implementand interpret score tests of normality in the innovations to the latent variables in state space modelsagainst generalized hyperbolic alternatives, including symmetric and asymmetric Student ts. Wedecompose our tests into third and fourth moment components, and obtain one-sided likelihood ratioanalogues, whose asymptotic distribution we provide. When we apply them to a cointegrateddynamic factor model which combines the expenditure and income versions of US aggregate realoutput to improve its measurement, we reject normality if the sample period extends beyond theGreat Moderation.
