General bootstrap for dual \(\phi\)-divergence estimates (Q764446): Difference between revisions

Summary: A general notion of bootstrapped \(\phi\)-divergence estimates constructed by an exchangeably weighting sample is introduced. Asymptotic properties of these generalized bootstrapped \(\phi\)-divergence estimates are obtained, by means of the empirical process theory, which are applied to construct a bootstrap confidence set with asymptotically correct coverage probability. Some practical problems are discussed, including, in particular, the choice of an escort parameter, and several examples of divergences are investigated. Simulation results are provided to illustrate the finite sample performance of the proposed estimators.