Searcharxiv⌕ Search

arXiv subjects

Isha Dewan

Publications and source records attributed to Isha Dewan.

6 recordsLinked to original sources
arXiv2025

Empirical Likelihood Based Inference for a Divergence Measure Based on Survival Extropy

Survival extropy, which quantifies the uncertainty associated with the remaining lifetime distribution, provides an information-theoretic perspective on survival behavior. We consider a divergence measure based on survival extropy and derive its nonparametric estimators based on U-statistics, empirical distribution functions, and kernel density. Further, we construct confidence intervals for the divergence measure using the jackknife empirical likelihood (JEL) method and the normal approximation method with a jackknife pseudo-value-based variance estimator. A comprehensive Monte Carlo simulation study is conducted to compare the performance of the measure with existing divergence measures. Additionally, we evaluate the finite-sample performance of various estimators for the proposed measure. The findings highlight the effectiveness of the divergence measure and its estimators in practical applications. Finally, we show how the proposed divergence measure is used to detect the small differences between images in image datasets.

math.ST↗
arXiv2024

Jackknife Empirical Likelihood Method for U Statistics Based on Multivariate Samples and its Applications

We develop a jackknife empirical likelihood (JEL) framework for inference on parameters defined through multivariate three-sample U-statistic. From three independent multivariate samples, we construct JEL ratio statistic based on suitable jackknife pseudo-values and, under mild regularity conditions, establish a Wilks-type result showing that the log JEL ratio converges in distribution to a chi-square limit. This provides asymptotically valid confidence intervals for the parameter of interest without explicit variance estimation or heavy resampling. To illustrate the usefulness of the proposed method, we construct confidence intervals for differences in volume under the surface (VUS) measures, which are widely used in classification problems. Through Monte Carlo simulations, we compare the performance of JEL-based confidence intervals with those obtained from normal approximation of U-statistic and kernel-based methods. The findings indicate that the proposed JEL approach outperforms existing methods in terms of coverage probability and computational efficiency. Finally, we apply our methods to a recent real dataset.

stat.ME↗
arXiv2021

A non-parametric test for testing independence between time to failure and cause of failure of discrete competing risks data

Competing risks data with discrete lifetime comes up in practice. However, only limited literature exists for such data. In this paper, we propose a non-parametric test based on U-statistics for testing independence of time to failure and cause of failure of competing risks data when the lifetime is a discrete random variable. Asymptotic distribution of the proposed test statistic is derived. An extensive Monte Carlo simulation study is conducted to assess the finite sample performance of the proposed test. The flexibility of the testing procedure is illustrated using real data sets on oral cancer patients and drug exposed pregnancies.

stat.ME↗
arXiv2018

Testing for exponentiality for stationary associated random variables

In this paper, we consider the problem of testing for exponentiality against univariate positive ageing when the underlying sample consists of stationary associated random variables. In particular, we discuss the asymptotic behavior of the tests by Deshpande (1983), Hollander and Proschan (1972) and Ahmad (1992) for testing exponentiality against IFRA, NBU and DMRL, respectively under association. A simulation study illustrates the effect of dependence on the asymptotic normality of the test statistics and on the size and power of the tests.

math.ST↗
arXiv2017

On limiting distribution of U-statistics based on associated random variables

Let $\{X_n, n \ge 1\}$ be a sequence of stationary associated random variables. We discuss another set of conditions under which a central limit theorem for U-statistics based on $\{X_n, n \ge 1\}$ holds. We look at U-statistics based on differentiable kernels of degree 2 and above. We also discuss some applications.

math.ST↗