Publicação
Essays on econometrics. Multivariate Markov chains
| Resumo: | This Dissertation is about Markov chains and their role in economics and in econometrics theory. Four essays on the Markov chain ap- proach are presented. We start by illustrating the analytical potential of multivariate Markov chains in the field of economic history, in particular with regard to a test of the Schumpeterian hypothesis of creative destruction. Then, we ilustrate the flexibility of Markov chains, and their per- tinence to situations that go beyond their traditional applicability: i) how can a Markov chain play the role of covariates; ii) how can a Markov chain representation be useful to compute expected hitting times. Finally, we present a new methodology for testing and detecting multiple structural breaks in multivariate Markov chains, where the dates at which the structural breaks occur are unknown. |
|---|---|
| Autores principais: | Damásio, Bruno |
| Assunto: | Cadeias de Markov Econometria Matemática aplicada à economia |
| Ano: | 2018 |
| País: | Portugal |
| Tipo de documento: | tese de doutoramento |
| Tipo de acesso: | acesso aberto |
| Instituição associada: | Universidade de Lisboa |
| Idioma: | inglês |
| Origem: | Repositório da Universidade de Lisboa |
| Resumo: | This Dissertation is about Markov chains and their role in economics and in econometrics theory. Four essays on the Markov chain ap- proach are presented. We start by illustrating the analytical potential of multivariate Markov chains in the field of economic history, in particular with regard to a test of the Schumpeterian hypothesis of creative destruction. Then, we ilustrate the flexibility of Markov chains, and their per- tinence to situations that go beyond their traditional applicability: i) how can a Markov chain play the role of covariates; ii) how can a Markov chain representation be useful to compute expected hitting times. Finally, we present a new methodology for testing and detecting multiple structural breaks in multivariate Markov chains, where the dates at which the structural breaks occur are unknown. |
|---|