Abstract
dc:description.abstractWe apply regime-switching models to study the dynamic switching behaviour of equity risk premia. Traditionally, equity risk premia have been estimated assuming a single regime exists. Regime-switching models allow for the existence of two, or more, regimes. Three regime-switching models are employed: structural break models, threshold models and Markov regime-switching models. Both structural break models and threshold models assume that the switching mechanism is deterministic. The former allow for only a single break and the state variable is solely determined by time. Under the latter, multiple changes are allowed and the state variable is determined by an observable variable with respect to an unobserved threshold. In Markov regime-switching models, equity risk premia are allowed to switch probabilistically for each observation. This is achieved by introducing a state variable which is governed by a Markov process. To capture the co-movements among financial variables, we extend regime-switching models to a VAR framework, employing threshold autoregressive vector models and Markov regime-switching vector models. We estimate models of UK equity risk premia conditionally on the state variable which is related to business conditions. The results of non-linearity tests favour regime-switching models and suggest that regime-switching is an important characteristic of UK equity risk premia.
Degree
thesis:*- Name dc:type.qualificationname
- d_ph
- Level dc:type.qualificationlevel
- d_ph
- Grantor dc:publisher.institution
- University of Birmingham
- Year dc:date.issued
- 2013
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Tan, Min
Subjects
dc:subject × 1Rights
- Language dc:language.dcterms_iso639_2
- eng