We derive new tests for proper calibration of multivariate density forecasts based on Rosenblatt probability integral transforms. These tests have the advantage that they i) do not depend on the ordering of variables in the forecasting model, ii) are applicable to densities of arbitrary dimensions, and iii) have superior power relative to existing approaches. We furthermore develop adjusted tests that allow for estimated parameters and, consequently, can be used as in-sample specification tests. We demonstrate the problems of existing tests and how our new approaches can overcome those using Monte Carlo Simulation as well as two applications based on multivariate GARCH-based models for stock market returns and on a macroeconomic Bayesian vectorautoregressive model.
density calibration goodness-of-fit test predictive density Rosenblatt transformation