Abstract
We employ a variables-in-common method for constructing multivariate Tweedie distributions, based on linear combinations of independent univariate Tweedie variables. The method lies on the convolution and scaling properties of the Tweedie laws, using the cumulant generating function for characterization of the distributions and correlation structure. The routine allows the equivalence between independence and zero correlation and gives a parametrization through given values of the mean vector and dispersion matrix, similarly to the Gaussian vector. Our approach leads to a matrix representation of multivariate Tweedie models, which permits the simulations of many known distributions, including Gaussian, Poisson, non-central gamma, gamma, and inverse Gaussian, both positively or negatively correlated.
Similar content being viewed by others
References
Boubacar Maïnassara Y, Kokonendji CC (2014) On normal stable Tweedie models and power-generalized variance functions of only one component. TEST 23:585–606
Devroye L (1986) Non-uniform random variate generation. Springer, New York
Dunn PK (2014) R package ‘Tweedie’ http://cran.r-project.org/web/packages/tweedie/tweedie.pdf
Edgeman RL (1989) Inverse Gaussian control charts. Aust J Stat 31:78–84
Foster SD, Bravington MV (2013) A Poisson-gamma model for analysis of ecological non-negative continuous data. Environ Ecol Stat 20:533–552
Furman E, Landsman Z (2007) Economic capital allocations for non-negative portfolios of dependent risks. Austin Bull 38:601–619
Furman E, Landsman Z (2010) Multivariate Tweedie distributions and some related capital-at-risk analysis. Insur Math Econ 46:143–158
Hassairi A, Louati M (2009) Multivariate stable exponential families and Tweedie scale. J Stat Plan Inference 139:143–158
Holgate P (1964) Estimation for the bivariate Poisson distribution. Biometrika 51:241–287
Joe H (1997) Multivariate models and dependence concepts. Chapman & Hall, London
Jørgensen B (1997) The theory of dispersion models. Chapman & Hall, London
Jørgensen B (2013) Construction of multivariate dispersion models. Braz J Probab Stat 27:285–309
Jørgensen B, Martínez JR (2013) Multivariate exponential dispersion models. In: Kollo T (ed) The IX Tartu conference on multivariate statistics & XX international workshop on matrices and statistics. Proceeding. World Scientific, Singapore
Jørgensen B, Kokonendji CC (2015) Discrete dispersion models and their Tweedie asymptotics. AStA Adv Stat Anal. doi:10.1007/s10182-015-0250-z
Kendal WS (2014) Multifractality attributed to dual central limit-like convergence effects. Phys A Stat Mech Appl 401:22–23
Kokonendji CC, Dossou-Gbété S, Demétrio CGB (2004) Some discrete exponential dispersion models: Poisson–Tweedie and Hinde–Demétrio classes. SORT 28:201–214
R Development Core Team (2014) R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna
Smyth KG, Jørgensen B (2002) Fitting Tweedie’s compound Poisson model to insurance claims data: dispersion modelling. Austin Bull 32:143–157
Tweedie MCK (1984) An index which distinguishes between some important exponential families. In: Ghosh JK, Roy J (eds) Statistics: applications and new directions. Proceeding; Indian Statistical Golden Jubilee International Conference, Calcutta, pp 579–604
Wheeler B (2009) R package ‘SuppDists’ http://cran.r-project.org/web/packages/SuppDists/SuppDists.pdf
Acknowledgments
The authors would like to sincerely thank the editor, the associate editor as well as the anonymous referees for their valuable comments and suggestions which led to significant improvement in this article.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Cuenin, J., Jørgensen, B. & Kokonendji, C.C. Simulations of full multivariate Tweedie with flexible dependence structure. Comput Stat 31, 1477–1492 (2016). https://doi.org/10.1007/s00180-015-0617-3
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00180-015-0617-3