A misspecification test for multiplicative error models of non-negative time series processes

A class of Minimum Distance Estimators in Markovian Multiplicative Error Models

“…These models have been found to be useful for modeling a variety of non-negative time series. Numerous applications and properties of these models are discussed in Engle and Russell (1998), Bauwens and Giot (2000), Bauwens and Veredas (2004), Manganelli (2005), Chou (2005), Engle and Gallo (2006), Fernandes and Grammig (2006), Gao et al (2015) and Koul and Perera (2021), among others.…”

Bootstrap based probability forecasting in multiplicative error models

“…In finance, the family of multiplicative error models plays a key role in modelling non-negative valued time series processes (Engle, 2002;Pacurar, 2008). For example, they have been used for modelling financial durations (Engle and Russell, 1998;Allen et al, 2008;Gao et al, 2015), trading volume of orders (Manganelli, 2005), high-low range of asset prices (Chou, 2005), spikes in electricity price (Christensen et al, 2012), absolute value of daily returns (Engle and Gallo, 2006), and realized volatility (Brownlees et al, 2012). This paper develops new methodology for producing multi-step ahead probability forecasts for a nonnegative valued time-series obeying a multiplicative error model.…”

“…The existing methods for forecasting in MEMs are mainly based on point forecasts (see Engle and Russell, 1997;Dufour and Engle, 2000;Bauwens and Giot, 2001;Hautsch, 2011;Luca et al, 2017) and evaluating one-step-ahead density forecasts (Bauwens et al, 2004;Corsi et al, 2008;Hautsch et al, 2014). Although such methods have been widely used in empirical studies (see Bauwens and Giot, 2000;Fernandes and Grammig, 2005;Engle and Gallo, 2006;Corsi et al, 2008;Gao et al, 2015 and references there in), the literature is scant on methods for interval forecasts or multistep-ahead density/distribution forecasts. However, in risk management, for example in managing financial investments, one needs to take into account of the forecast of not only the very next observation, but also of those observations that are several steps ahead.…”

Simultaneous Diagnostic Testing for Nonlinear Time Series Models with An Application to the U.S. Federal Fund Rate

“…(), Gao and Casas (), among others. On the other hand, Horváth and Zitikis () and Koul and Ling () develop specification tests for the error distribution in GARCH type models and Gao, Kim and Saart () consider specification testing for the error distribution in the multiplicative error model.…”