Marra, G. and Radice, Rosalba (2017) Bivariate copula additive models for location, scale and shape. Computational Statistics and Data AnalysiS 112 , pp. 99-113. ISSN 0167-9473.
|
Text
copulaReg.pdf - Author's Accepted Manuscript Download (766kB) | Preview |
Abstract
n generalized additive models for location, scale and shape (GAMLSS), the response dis- tribution is not restricted to belong to the exponential family and all the model’s pa rameters can be made dependent on additive predictors that allow for several typ es of covariate effects (such as linear, non-linear, random and spatial effects). In many empir ical situations, how- ever, modeling simultaneously two or more responses conditional on some cov ariates can be of considerable relevance. The scope of GAMLSS is extended by introd ucing bivariate cop- ula models with continuous margins for the GAMLSS class. The proposed comp utational tool permits the copula dependence and marginal distribution parameters to be estimated si- multaneously, and each parameter to be modeled using an additive predictor. Simultaneous parameter estimation is achieved within a penalized likelihood framework using a tr ust region algorithm with integrated automatic multiple smoothing parameter selection. The introdu ced approach allows for straightforward inclusion of potentially any parametric marginal distribu- tion and copula function. The models can be easily used via the copulaReg() function in the R package SemiParBIVProbit . The proposal is illustrated through two case studies and simulated data.
Metadata
| Item Type: | Article |
|---|---|
| Keyword(s) / Subject(s): | additive predictor, marginal distribution, copula, simultaneous parameter estimation. |
| School: | Birkbeck Faculties and Schools > Faculty of Business and Law > Birkbeck Business School |
| Depositing User: | Rosalba Radice |
| Date Deposited: | 03 Mar 2017 15:18 |
| Last Modified: | 02 Aug 2023 17:31 |
| URI: | https://eprints.bbk.ac.uk/id/eprint/18257 |
Statistics
Additional statistics are available via IRStats2.
Archive Staff Only (login required)
