A Multivariate Generalized Orthogonal Factor GARCH Model
Bibliographic Data
| ID | 19420201 |
|---|---|
| Authors | Markku Lanne (0000-0001-9397-2578, University of Helsinki), Pentti Saikkonen (Statistics Finland) |
| Year | 2007 |
| Volume | 25 |
| Issue | 1 |
| Pages | 61-75 |
| Publication date | 2007-01-01 |
| Open Access | No |
| Type | ARTICLE |
| Venue | Journal of Business and Economic Statistics (JOURNAL) |
| Journal identifiers | ISSN: 0735-0015 • E-ISSN: 1537-2707 |
| Publisher | Informa UK Limited (PUBLISHER • GB) |
| DOI | 10.1198/073500106000000404 |
| OpenAlex | W1512502159 |
| Language | EN |
| Citations received | 4 |
| References cited | 15 |
We propose a factor generalized autoregressive conditional heteroscedasticity (GARCH) model and develop test procedures for checking the correctness of the number of factors. Maximum likelihood estimation of the model is straightforward once computationally simple preliminary estimates are obtained. Motivated by the empirical application of the article, a mixture of Gaussian distributions is considered in addition to the conventional Gaussian likelihood. Interestingly, some parameters of the conditional covariance matrix that are not identifiable under normality, can be identified when the mixture specification is used. As an empirical example, modeling a system of exchange rate returns and testing for volatility transmission is considered
Autoregressive conditional heteroskedasticity · Conditional variance · Covariance · Covariance matrix · Econometrics · Factor analysis · Heteroscedasticity · Likelihood function · Likelihood-ratio test · Maximum likelihood · Multivariate normal distribution · Multivariate statistics · Statistics · Financial Risk and Volatility Modeling · Mathematics · Applied Mathematics
| Citations per year | 0.21 |
|---|---|
| Citation span | 2007 - 2022 (16) |
| Highly cited | No |