Refine
Document Type
- Report (5)
Language
- English (5)
Has Fulltext
- yes (5)
Is part of the Bibliography
- no (5)
Keywords
Institute
- Wirtschaftswissenschaften (5) (remove)
We propose a new framework for modelling the time dependence in duration processes being in force on financial markets. The pioneering ACD model introduced by Engle and Russell (1998) will be extended in a manner that the duration process will be accompanied by an unobservable stochastic process. The Discrete Mixture ACD framework provides us with a general methodology which puts the idea into practice. It is established by introducing a discrete-valued latent regime variable which can be justified in the light of recent market microstructure theories. The empirical application demonstrates its ability to capture specific characteristics of intraday transaction durations while alternative approaches fail. JEL classification: C41, C22, C25, C51, G14.
In recent methodological work the well known ACD approach, originally introduced by Engle and Russell (1998), has been supplemented by the involvement of an unobservable stochastic process which accompanies the underlying process of durations via a discrete mixture of distributions. The Mixture ACD model, emanating from the specialized proposal of De Luca and Gallo (2004), has proved to be a moderate tool for description of financial duration data. The use of one and the same family of ordinary distributions has been common practice until now. Our contribution incites to use the rich parameterized comprehensive family of distributions which allows for interacting different distributional idiosyncrasies. JEL classification: C41, C22, C25, C51, G14.
We propose a new framework for modelling the time dependence in duration processes being in force on financial markets. The pioneering ACD model introduced by Engle and Russell (1998) will be extended in a manner that the duration process will be accompanied by an unobservable stochastic process. The Discrete Mixture ACD framework provides us with a general methodology which puts the idea into practice. It is established by introducing a discrete-valued latent regime variable which can be justified in the light of recent market microstructure theories. The empirical application demonstrates its ability to capture specific characteristics of intraday transaction durations while alternative approaches fail. JEL classification: C41, C22, C25, C51, G14.
In recent methodological work the well known ACD approach, originally introduced by Engle and Russell (1998), has been supplemented by the involvement of an unobservable stochastic process which accompanies the underlying process of durations via a discrete mixture of distributions. The Mixture ACD model, emanating from the specialized proposal of De Luca and Gallo (2004), has proved to be a moderate tool for description of financial duration data. The use of one and the same family of ordinary distributions has been common practice until now. Our contribution incites to use the rich parameterized comprehensive family of distributions which allows for interacting different distributional idiosyncrasies. JEL classification: C41, C22, C25, C51, G14
We propose a new framework for modeling time dependence in duration processes. The ACD approach introduced by Engle and Russell (1998) will be extended so that the conditional expectation of the durations depends on an unobservable stochastic process which is modeled via a Markov chain. The Markov switching ACD model (MSACD) is a flexible tool for description of financial duration processes. The introduction of a latent information regime variable can be justified in the light of recent market microstructure theories. In an empirical application we show that the MSACD approach is able to capture specific characteristics of inter trade durations while alternative ACD models fail. JEL classification: C41, C22, C25, C51, G14