Research Commons
      • Browse 
        • Communities & Collections
        • Titles
        • Authors
        • By Issue Date
        • Subjects
        • Types
        • Series
      • Help 
        • About
        • Collection Policy
        • OA Mandate Guidelines
        • Guidelines FAQ
        • Contact Us
      • My Account 
        • Sign In
        • Register
      View Item 
      •   Research Commons
      • University of Waikato Research
      • Science and Engineering
      • Science and Engineering Papers
      • View Item
      •   Research Commons
      • University of Waikato Research
      • Science and Engineering
      • Science and Engineering Papers
      • View Item
      JavaScript is disabled for your browser. Some features of this site may not work without it.

      Line mesh distributions: An alternative approach for multivariate environmental extremes

      Bardsley, W. Earl; Vetrova, Varvara; Dao, Ngoc Hieu
      DOI
       10.1007/s00477-018-1642-x
      Link
       rdcu.be
      Find in your library  
      Citation
      Export citation
      Bardsley, W. E., Vetrova, V., & Dao, N. H. (2019). Line mesh distributions: An alternative approach for multivariate environmental extremes. Stochastic Environmental Research and Risk Assessment. https://doi.org/10.1007/s00477-018-1642-x
      Permanent Research Commons link: https://hdl.handle.net/10289/12251
      Abstract
      Copulas and other multivariate models can give joint exceedance probabilities for multivariate events in the naturalenvironment. However, the choice of the most appropriate multivariate model may not always be evident in the absence ofknowledge of dependence structures. A simple nonparametric alternative is to approximate multivariate dependenciesusing ‘‘line mesh distributions’’, introduced here as a data-based finite mixture of univariate distributions defined on a meshof L =C(m, 2) lines extending through Euclidean n-space. That is, m data points in n-space define a total of L lines, whereC() denotes the binomial coefficient. The utilitarian simplicity of the method has attraction for joint exceedance proba-bilities because just the data and a single bandwidth parameter within the 0, 1 interval are needed to define a line meshdistribution. All bivariate planes in these distributions have the same Pearson correlation coefficients as the correspondingdata. Marginal means and variances are similarly preserved. Using an example from the literature, a 5-parameter bivariateGumbel model is replaced with a 1-parameter line mesh distribution. A second illustration for three dimensions applies linemesh distributions to data simulated from a trivariate copula.
      Date
      2019
      Type
      Journal Article
      Publisher
      Springer (part of Springer Nature)
      Collections
      • Science and Engineering Papers [3086]
      Show full item record  

      Usage

       
       
       

      Usage Statistics

      For this itemFor all of Research Commons

      The University of Waikato - Te Whare Wānanga o WaikatoFeedback and RequestsCopyright and Legal Statement