We propose an extension of the Generalized Autocontour (G-ACR) tests for dynamic specification of in-sample conditional densities and for evaluation of out-of-sample forecast densities. The new tests are based on probability integral transforms (PITs) computed from bootstrap conditional densities that incorporate parameter uncertainty. Then, the parametric specification of the conditional moments * Financial support from the Spanish Ministry of Education and Science, research project ECO2015-70331-C2-2-R (MINECO/FEDER) is acknowledged by the four authors. The fourth author also acknowledges research project ECO2012-3240 and FCT grant UID/GES/00315/2013. Gloria González-Rivera wishes to thank the Department of Statistics at UC3M for their hospitality and the financial support of the 2015 Chair of Excellence UC3M/Banco de Santander, and the UCR Academic Senate grant. We are grateful to the participants at the New Developments in Econometrics and Time Series Workshop, Madrid, October 2016, and at the IMF/IIF Workshop on Forecasting Issues on Developing Economies, Washington DC, April 2017, and to seminar participants at the Management School of the University of Liverpool, for their very useful comments. We are also thankful to J. Mencía and E. Sentana for their help with the codes to estimate the MEM model.
1can be tested without relying on any parametric error distribution yet exploiting distributional properties of the variable of interest. We show that the finite sample distribution of the bootstrapped G-ACR (BG-ACR) tests are well approximated using standard asymptotic distributions. Furthermore, the proposed tests are easy to implement and are accompanied by graphical tools that provide information about the potential sources of misspecification. We apply the BG-ACR tests to the Heterogeneous Autoregressive (HAR) model and the Multiplicative Error Model (MEM) of the U.S. volatility index VIX. We find strong evidence against the parametric assumptions of the conditional densities, i.e. normality in the HAR model and semi non-parametric Gamma (GSNP) in the MEM. In both cases, the true conditional density seems to be more skewed to the right and more peaked than either normal or GSNP densities, with location, variance and skewness changing over time. The preferred specification is the heteroscedastic HAR model with bootstrap conditional densities of the log-VIX. Supplementary materials for this article are available online.