Estimating a multivariate model with discrete Weibull margins
Recently, a proposal for simulating correlated discrete Weibull variables has been suggested, based on the Gaussian copula. Although the procedure is straightforward and allows the user to directly assign the desired pairwise correlations between the discrete margins (or simply assign the correlation matrix of the Gaussian copula), on the other hand the estimation process is more tricky. In this work, we describe and assess, with a special focus on the bivariate case, three techniques for the estimation of the model parameters; an application to real data is provided as well.
KeywordsCorrelated counts Gaussian copula inference function for margins maximum likelihood
AMS Subject Classification62H20 65C60
Unable to display preview. Download preview PDF.
- Barbiero, A. 2015a. DiscreteWeibull: Discrete Weibull distributions (Type 1 and 3), R package version 1.1. https://doi.org/CRAN.R-project.org/package=DiscreteWeibull.
- Barbiero, A. 2015b. Simulating correlated discrete Weibull variables: a proposal and an implementation in the R environment. In International Conference of Computational Methods in Science and Engineering 2015, AIP conference proceedings 1702, eds. T. E. Simos, Z. Kalogiratou, and T. Monovasilis, 190017.Google Scholar
- Barbiero, A., and P. A. Ferrari. 2015a. GenOrd: Simulation of ordinal and discrete variables with given correlation matrix and marginal distributions, R package version 1.4.0. https://doi.org/CRAN.R-project.org/package=GenOrd.
- Cario, M. C., and B. L. Nelson. 1997. Modeling and generating random vectors with arbitrary marginal distributions and correlation matrix. Technical report, Department of Industrial Engineering and Management Sciences, Northwestern University, Evanston, IL.Google Scholar
- R Core Team 2016. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. https://doi.org/www.R-project.org.