Scholarlink

<p>The exponential distribution is one of the most widely used continuous distributions in reliability theory, failure time analysis, Poisson processes, and survival analysis. In this paper, we establish new characterizations of the exponential distribution based on the probability integral transformation. Characterizations derived from both distributional relationships and moment properties of certain functions of the underlying variables are presented. Furthermore, analogous results for distributional relationships are obtained for functions of order statistics and <i>k</i>-record statistics. Using the completeness properties of the sequences of functions {(1 − <i>u</i>)<sup><i>i</i>−1</sup><i>u</i><sup><i>n</i>−<i>i</i></sup>, 0 &lt; <i>u</i> &lt; 1, <i>n</i> ⩾ <i>i</i> ⩾ 1} and {(− log <i>u</i>)<sup><i>n</i>−1</sup>, 0 &lt; <i>u</i> &lt; 1, <i>n</i> ⩾ 1}, moment-based characterization results are also established for functions of order statistics and <i>k</i>-record statistics, respectively. Since ordinary record values are special cases of <i>k</i>-records, the corresponding results hold for usual records as well.</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

Symmetry-based characterizations of the exponential distribution

  • 期刊论文

  • 发表日期：

    2026年6月30日

查看全文
  • Jafar Ahmadi,
  • Massoumeh Fashandi

摘要

The exponential distribution is one of the most widely used continuous distributions in reliability theory, failure time analysis, Poisson processes, and survival analysis. In this paper, we establish new characterizations of the exponential distribution based on the probability integral transformation. Characterizations derived from both distributional relationships and moment properties of certain functions of the underlying variables are presented. Furthermore, analogous results for distributional relationships are obtained for functions of order statistics and k-record statistics. Using the completeness properties of the sequences of functions {(1 − u)i−1un−i, 0 < u < 1, n ⩾ i ⩾ 1} and {(− log u)n−1, 0 < u < 1, n ⩾ 1}, moment-based characterization results are also established for functions of order statistics and k-record statistics, respectively. Since ordinary record values are special cases of k-records, the corresponding results hold for usual records as well.