Multivariate L-moment based tests for copula selection, with hydrometeorological applications
2019
Ben Nasr, I. | Chebana, F.
Hydrological and climatological extreme events are characterized by several correlated random variables. For a better associated risk assessment, the dependence structure between these variables must be taken into account by considering copulas. Multiparameter copulas (M-copulas) play an important role by their flexibility and ability to capture more than one type of dependence. Since model misspecification can lead to underestimation or overestimation of the associated risk, it is of high importance to select the appropriate copula. To this end, several goodness-of-fit (GOF) tests have been proposed. However, these tests are applied, validated and evaluated for the usual one-parameter copulas. Nevertheless, there is no specific GOF test for M-copulas that takes into account the specificities of hydrometeorological series. Hence, the aim of the present paper is to introduce new GOF tests specifically for M-copulas and adapted to hydrometeorological context. More precisely, the proposed GOF tests are based on multivariate L-moments. A simulation study is conducted to evaluate and compare the performances of the proposed tests. The results confirm the usefulness of the new GOF tests in comparison with some well-established ones. Finally, the newly introduced tests are illustrated on hydrometeorological data series.
Show more [+] Less [-]AGROVOC Keywords
Bibliographic information
This bibliographic record has been provided by National Agricultural Library