While theory of autoregressive conditional heteroskedasticity (ARCH) models is well understood for strictly stationary processes, some recent interest has focused on the nonstationary case. In the classical model including a positive intercept parameter, the volatility process diverges to...

The Tobit model (censored regression model) is an important basic model appearing in many applications in economics. In this paper we consider a duration Tobit model in which a duration variable which counts the number of times the data is being censored is included as a covariate. We show that...

A Monte Carlo approach is suggested for correctly sized backtesting of Value-at-Risk estimates by means of the dynamic quantile test and a Portmanteau statistic. The latter shows preferable power features but fails in case of unconditional VaR misspecification.

We test for causality between inflation and its associated uncertainty by means of both in-sample and out-of-sample modelling. Our findings indicate that the impact of inflation on inflation uncertainty is more pronounced than the reverse causal effect.