Scholarlink

<p>We propose a non-central formulation of the multivariate skew <i>t</i> (NCST) distribution, defined by scaling skew normal random vectors by independent chi-squared variables. This construction extends the classical multivariate <i>t</i> family by allowing both asymmetry and non-centrality, which provides an alternative to existing skew <i>t</i> models that often rely on restrictive assumptions for tractability. We derive key theoretical properties of the NCST distribution, including its moment structure, behavior under affine transformations, and the distribution of quadratic forms. To demonstrate practical utility, we apply the NCST model to breast cancer diagnostic data, modeling multiple features of tumor shape. The NCST model achieves a superior fit, as indicated by information criteria and visual diagnostics, particularly in the presence of skewness and heavy tails, compared to standard alternatives, including the multivariate normal, skew normal, and Azzalini’s skew <i>t</i> distributions. Our findings suggest that the NCST distribution offers a useful and interpretable choice for modeling complex multivariate data, which highlights promising directions for future development in likelihood inference, Bayesian computation, and applications involving asymmetry and non-Gaussian dependence.</p>

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

Flexible Modeling of Multivariate Skewed and Heavy-Tailed Data via a Non-Central Skew t Distribution with Application to Tumor Shape Data

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年5月19日

查看全文
  • Abeer M. Hasan,
  • Ying-Ju Chen

摘要

We propose a non-central formulation of the multivariate skew t (NCST) distribution, defined by scaling skew normal random vectors by independent chi-squared variables. This construction extends the classical multivariate t family by allowing both asymmetry and non-centrality, which provides an alternative to existing skew t models that often rely on restrictive assumptions for tractability. We derive key theoretical properties of the NCST distribution, including its moment structure, behavior under affine transformations, and the distribution of quadratic forms. To demonstrate practical utility, we apply the NCST model to breast cancer diagnostic data, modeling multiple features of tumor shape. The NCST model achieves a superior fit, as indicated by information criteria and visual diagnostics, particularly in the presence of skewness and heavy tails, compared to standard alternatives, including the multivariate normal, skew normal, and Azzalini’s skew t distributions. Our findings suggest that the NCST distribution offers a useful and interpretable choice for modeling complex multivariate data, which highlights promising directions for future development in likelihood inference, Bayesian computation, and applications involving asymmetry and non-Gaussian dependence.