Abstract
Signals coming from multivariate higher-order conditional moments as well as the information contained in exogenous covariates can be exploited by rational investors to allocate their wealth among different risky investment opportunities. This paper proposes a new flexible dynamic copula model being able to explain and forecast the time-varying shape of large dimensional asset returns distributions. The time-varying dependence is introduced by allowing the dynamic updating equation of the copula correlation parameters to depend on a latent Markov-switching process as well as on exogenous covariates. As a further key ingredient of the model specification, we let the univariate marginals to be driven by an updating mechanism based on the scaled score of the conditional distribution. This framework allows us to introduce time-variation in the conditional moments up to the fourth order. Time-varying moments are then used to build a portfolio allocation strategy that maximises the utility function of a representative rational investor. We empirically assess that the proposed model substantially improves the optimal portfolio allocation with respect to competing alternative investment strategies.
| Original language | English |
|---|---|
| Journal | Journal of Empirical Finance |
| Volume | 48 |
| Issue | September |
| Pages (from-to) | 1-18 |
| Number of pages | 18 |
| ISSN | 0927-5398 |
| DOIs | |
| Publication status | Published - 2018 |
Keywords
- Dynamic conditional score
- Dynamic copula
- Generalized autoregressive score
- Higher order moments
- Markov-switching
- Portfolio optimisation