Scholarlink

In this chapter, we consider the approximation of sample copulas and propose a strategy for model selection of copulas obtained as an approximation. For this reason, we introduce the weighted Cramér von Mises divergence with regularization term. This term is added to prevent overfitting. The unknown copula parameters of interest are estimated by an approximate minimum-distance estimator. A consistency result and a theorem about the asymptotic normality of this estimator are provided. Furthermore, we establish tests for comparisons of several models, including a theorem about the asymptotic distribution of the test statistic.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Selection of Parametric Copula Models in the Approximation of Copulas Using Cramér-von Mises Divergence

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Eckhard Liebscher

摘要

In this chapter, we consider the approximation of sample copulas and propose a strategy for model selection of copulas obtained as an approximation. For this reason, we introduce the weighted Cramér von Mises divergence with regularization term. This term is added to prevent overfitting. The unknown copula parameters of interest are estimated by an approximate minimum-distance estimator. A consistency result and a theorem about the asymptotic normality of this estimator are provided. Furthermore, we establish tests for comparisons of several models, including a theorem about the asymptotic distribution of the test statistic.