Pub Date : 2025-06-30DOI: 10.1016/j.jspi.2025.106315
Tommaso Lando, Sirio Legramanti
Given a pair of non-negative random variables and , we introduce a class of nonparametric tests for the null hypothesis that dominates in the total time on test order. Critical values are determined using bootstrap-based inference, and the tests are shown to be consistent. The same approach is used to construct tests for the excess wealth order. As a byproduct, we also obtain a class of goodness-of-fit tests for the NBUE (New Better than Used in Expectation) family of distributions.
给定一对非负随机变量X和Y，我们引入了对X在总时间上优于Y的零假设的一类非参数检验。使用基于自举的推理来确定临界值，并且测试表明是一致的。同样的方法也用于构造过剩财富顺序的检验。作为副产品，我们还获得了NBUE （New Better than Used in Expectation）分布族的一类拟合优度检验。
{"title":"Bootstrap-based tests for the total time on test and the excess wealth orders","authors":"Tommaso Lando, Sirio Legramanti","doi":"10.1016/j.jspi.2025.106315","DOIUrl":"10.1016/j.jspi.2025.106315","url":null,"abstract":"<div><div>Given a pair of non-negative random variables <span><math><mi>X</mi></math></span> and <span><math><mi>Y</mi></math></span>, we introduce a class of nonparametric tests for the null hypothesis that <span><math><mi>X</mi></math></span> dominates <span><math><mi>Y</mi></math></span> in the total time on test order. Critical values are determined using bootstrap-based inference, and the tests are shown to be consistent. The same approach is used to construct tests for the excess wealth order. As a byproduct, we also obtain a class of goodness-of-fit tests for the NBUE (New Better than Used in Expectation) family of distributions.</div></div>","PeriodicalId":50039,"journal":{"name":"Journal of Statistical Planning and Inference","volume":"241 ","pages":"Article 106315"},"PeriodicalIF":0.8,"publicationDate":"2025-06-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144523736","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-06-29DOI: 10.1016/j.jspi.2025.106314
Ye Tian , Xinwei Zhang , Zhiqiang Tan
Consider a semi-supervised setting with a labeled dataset of binary responses and predictors and an unlabeled dataset with only the predictors. Logistic regression is equivalent to an exponential tilt model in the labeled population. For semi-supervised estimation of regression coefficients in logistic regression, we develop further analysis and understanding of a statistical approach using exponential tilt mixture (ETM) models and maximum nonparametric likelihood estimation, while allowing that the class proportions may differ between the unlabeled and labeled data. We derive asymptotic properties of ETM-based estimation and demonstrate improved efficiency over supervised logistic regression in a random sampling setup and an outcome-stratified sampling setup previously used. Moreover, we reconcile such efficiency improvement with the existing semiparametric efficiency theory when the class proportions in the unlabeled and labeled data are restricted to be the same. We also provide a simulation study to numerically illustrate our theoretical findings.
{"title":"On semi-supervised estimation using exponential tilt mixture models","authors":"Ye Tian , Xinwei Zhang , Zhiqiang Tan","doi":"10.1016/j.jspi.2025.106314","DOIUrl":"10.1016/j.jspi.2025.106314","url":null,"abstract":"<div><div>Consider a semi-supervised setting with a labeled dataset of binary responses and predictors and an unlabeled dataset with only the predictors. Logistic regression is equivalent to an exponential tilt model in the labeled population. For semi-supervised estimation of regression coefficients in logistic regression, we develop further analysis and understanding of a statistical approach using exponential tilt mixture (ETM) models and maximum nonparametric likelihood estimation, while allowing that the class proportions may differ between the unlabeled and labeled data. We derive asymptotic properties of ETM-based estimation and demonstrate improved efficiency over supervised logistic regression in a random sampling setup and an outcome-stratified sampling setup previously used. Moreover, we reconcile such efficiency improvement with the existing semiparametric efficiency theory when the class proportions in the unlabeled and labeled data are restricted to be the same. We also provide a simulation study to numerically illustrate our theoretical findings.</div></div>","PeriodicalId":50039,"journal":{"name":"Journal of Statistical Planning and Inference","volume":"241 ","pages":"Article 106314"},"PeriodicalIF":0.8,"publicationDate":"2025-06-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144549187","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-06-23DOI: 10.1016/j.jspi.2025.106313
Wenqing Su , Xiao Guo , Ying Yang
Multi-layer networks arise naturally in various domains including biology, finance and sociology, among others. The multi-layer stochastic block model (multi-layer SBM) is commonly used for community detection in the multi-layer networks. Most of current literature focuses on statistical consistency of community detection methods under multi-layer SBMs. However, the asymptotic distributional properties are also indispensable which play an important role in statistical inference. In this work, we aim to study the estimation and asymptotic properties of the layer-wise scaled connectivity matrices in the multi-layer SBM. We study and analyze a computationally tractable method for estimating the scaled connectivity matrices. Under the multi-layer SBM and its variant multi-layer degree-corrected SBM, we establish the asymptotic normality of the estimated matrices under mild conditions, which can be used for interval estimation and hypothesis testing. Simulations show the superior performance of proposed method over existing methods in two considered statistical inference tasks. We apply the method to a real dataset and obtain interpretable results. In addition, we develop a moment estimator for the non-scaled connectivity matrices and study its asymptotic properties.
{"title":"Limit results for estimation of connectivity matrix in multi-layer stochastic block models","authors":"Wenqing Su , Xiao Guo , Ying Yang","doi":"10.1016/j.jspi.2025.106313","DOIUrl":"10.1016/j.jspi.2025.106313","url":null,"abstract":"<div><div>Multi-layer networks arise naturally in various domains including biology, finance and sociology, among others. The multi-layer stochastic block model (multi-layer SBM) is commonly used for community detection in the multi-layer networks. Most of current literature focuses on statistical consistency of community detection methods under multi-layer SBMs. However, the asymptotic distributional properties are also indispensable which play an important role in statistical inference. In this work, we aim to study the estimation and asymptotic properties of the layer-wise scaled connectivity matrices in the multi-layer SBM. We study and analyze a computationally tractable method for estimating the scaled connectivity matrices. Under the multi-layer SBM and its variant multi-layer degree-corrected SBM, we establish the asymptotic normality of the estimated matrices under mild conditions, which can be used for interval estimation and hypothesis testing. Simulations show the superior performance of proposed method over existing methods in two considered statistical inference tasks. We apply the method to a real dataset and obtain interpretable results. In addition, we develop a moment estimator for the non-scaled connectivity matrices and study its asymptotic properties.</div></div>","PeriodicalId":50039,"journal":{"name":"Journal of Statistical Planning and Inference","volume":"241 ","pages":"Article 106313"},"PeriodicalIF":0.8,"publicationDate":"2025-06-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144500917","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-06-17DOI: 10.1016/j.jspi.2025.106310
Hanxiao Jing , Mary C. Meyer , Jiayang Sun
A straight-forward solution to the deconvolution density estimation involves penalized splines. A priori information about shape of the densities is readily imposed; for example the estimates may be constrained to be unimodal or bimodal. With quadratic splines and uniform errors, a cube-root convergence rate is attained. Simulations show that the estimators perform well compared to kernel estimators in a variety of scenarios.
{"title":"Deconvolution density estimation using penalized splines","authors":"Hanxiao Jing , Mary C. Meyer , Jiayang Sun","doi":"10.1016/j.jspi.2025.106310","DOIUrl":"10.1016/j.jspi.2025.106310","url":null,"abstract":"<div><div>A straight-forward solution to the deconvolution density estimation involves penalized splines. A priori information about shape of the densities is readily imposed; for example the estimates may be constrained to be unimodal or bimodal. With quadratic splines and uniform errors, a cube-root convergence rate is attained. Simulations show that the estimators perform well compared to kernel estimators in a variety of scenarios.</div></div>","PeriodicalId":50039,"journal":{"name":"Journal of Statistical Planning and Inference","volume":"241 ","pages":"Article 106310"},"PeriodicalIF":0.8,"publicationDate":"2025-06-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144366122","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-05-27DOI: 10.1016/j.jspi.2025.106297
Jin-Jian Hsieh, Siang-Ying Chen
This paper delves into the accelerated failure time model within the framework of dependent truncation data and leverages the copula model to establish correlations within the dataset. Building upon the work of Chaieb et al. (2006), who utilized the copula-graphic method to estimate survival functions and proposed an approach for estimating correlation parameters, we further extend the methodology by introducing two distinct estimation techniques for regression parameters. The first method involves parameter evaluation through the calculation of the area between survival curves, while the second method employs the weight of survival jump in conjunction with the least squares approach to estimate regression parameters. We evaluate the efficacy of these proposed estimation procedures through simulation studies and conduct a comparative analysis between the two approaches. Furthermore, we apply these methodologies to two real-world datasets, providing insights into their practical applicability. Through this analysis, we gain a deeper understanding of how these approaches can be effectively utilized in real-world scenarios.
{"title":"Accelerated failure time model under dependent truncated data","authors":"Jin-Jian Hsieh, Siang-Ying Chen","doi":"10.1016/j.jspi.2025.106297","DOIUrl":"10.1016/j.jspi.2025.106297","url":null,"abstract":"<div><div>This paper delves into the accelerated failure time model within the framework of dependent truncation data and leverages the copula model to establish correlations within the dataset. Building upon the work of Chaieb et al. (2006), who utilized the copula-graphic method to estimate survival functions and proposed an approach for estimating correlation parameters, we further extend the methodology by introducing two distinct estimation techniques for regression parameters. The first method involves parameter evaluation through the calculation of the area between survival curves, while the second method employs the weight of survival jump in conjunction with the least squares approach to estimate regression parameters. We evaluate the efficacy of these proposed estimation procedures through simulation studies and conduct a comparative analysis between the two approaches. Furthermore, we apply these methodologies to two real-world datasets, providing insights into their practical applicability. Through this analysis, we gain a deeper understanding of how these approaches can be effectively utilized in real-world scenarios.</div></div>","PeriodicalId":50039,"journal":{"name":"Journal of Statistical Planning and Inference","volume":"240 ","pages":"Article 106297"},"PeriodicalIF":0.8,"publicationDate":"2025-05-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144166460","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-05-25DOI: 10.1016/j.jspi.2025.106307
Aman Ullah , Tao Wang
We in this paper propose a stepwise estimation procedure for semiparametric modal regression with varying coefficients and measurement error, where the linear covariate is unobserved but an ancillary variable is available. This modal regression framework, which is built on the mode value rather than the mean, captures the “most likely” effect instead of the traditional average effect. The proposed stepwise procedure introduces a restricted regression mode by imposing a structural constraint on the model, allowing us to concentrate out the varying coefficients using the “correction for attenuation” method commonly employed in mean regression. This transformation reduces the original model to a parametric modal regression. We establish the consistency and asymptotic normality of the resulting modal estimators by analyzing the tail behavior of the characteristic function of the error distribution, distinguishing between ordinary smooth and super smooth cases. Additionally, we investigate bandwidth selection strategies and propose a simulation-extrapolation algorithm as a practical approach for optimal bandwidth choice. We conduct Monte Carlo simulations to assess the finite sample performance of the resulting estimators and present a real data analysis to further illustrate the effectiveness of the suggested estimation procedure.
{"title":"Semiparametric modal regression with varying coefficients and measurement error","authors":"Aman Ullah , Tao Wang","doi":"10.1016/j.jspi.2025.106307","DOIUrl":"10.1016/j.jspi.2025.106307","url":null,"abstract":"<div><div>We in this paper propose a stepwise estimation procedure for semiparametric modal regression with varying coefficients and measurement error, where the linear covariate is unobserved but an ancillary variable is available. This modal regression framework, which is built on the mode value rather than the mean, captures the “most likely” effect instead of the traditional average effect. The proposed stepwise procedure introduces a restricted regression mode by imposing a structural constraint on the model, allowing us to concentrate out the varying coefficients using the “correction for attenuation” method commonly employed in mean regression. This transformation reduces the original model to a parametric modal regression. We establish the consistency and asymptotic normality of the resulting modal estimators by analyzing the tail behavior of the characteristic function of the error distribution, distinguishing between ordinary smooth and super smooth cases. Additionally, we investigate bandwidth selection strategies and propose a simulation-extrapolation algorithm as a practical approach for optimal bandwidth choice. We conduct Monte Carlo simulations to assess the finite sample performance of the resulting estimators and present a real data analysis to further illustrate the effectiveness of the suggested estimation procedure.</div></div>","PeriodicalId":50039,"journal":{"name":"Journal of Statistical Planning and Inference","volume":"240 ","pages":"Article 106307"},"PeriodicalIF":0.8,"publicationDate":"2025-05-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144134836","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-05-19DOI: 10.1016/j.jspi.2025.106300
Zhen Zeng , Weixin Yao
When the size of the dataset becomes extremely large, it is computationally challenge for traditional statistical estimation methods and might be infeasible to store all the data on a single computer. Under the massive data framework, we extend the divide and conquer method to the generalized approximately expectile regression and investigate both of their finite and asymptotic properties. Bahadur representation of the estimators are established. Moreover, we prove that with the appropriate number of subsamples, the proposed method can ensure the accuracy of statistical inference. Simulations studies validate our theoretical findings. Supplementary materials for this article are available online.
{"title":"Divide and conquer for generalized approximately expectile regression","authors":"Zhen Zeng , Weixin Yao","doi":"10.1016/j.jspi.2025.106300","DOIUrl":"10.1016/j.jspi.2025.106300","url":null,"abstract":"<div><div>When the size of the dataset becomes extremely large, it is computationally challenge for traditional statistical estimation methods and might be infeasible to store all the data on a single computer. Under the massive data framework, we extend the divide and conquer method to the generalized approximately expectile regression and investigate both of their finite and asymptotic properties. Bahadur representation of the estimators are established. Moreover, we prove that with the appropriate number of subsamples, the proposed method can ensure the accuracy of statistical inference. Simulations studies validate our theoretical findings. Supplementary materials for this article are available online.</div></div>","PeriodicalId":50039,"journal":{"name":"Journal of Statistical Planning and Inference","volume":"240 ","pages":"Article 106300"},"PeriodicalIF":0.8,"publicationDate":"2025-05-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144139243","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-05-16DOI: 10.1016/j.jspi.2025.106299
Francesco Gili, Geurt Jongbloed, Aad van der Vaart
We consider nonparametric estimation of the distribution function of squared sphere radii in the classical Wicksell problem. Under smoothness conditions on in a neighborhood of , in Gili et al. (2024) it is shown that the Isotonic Inverse Estimator (IIE) is asymptotically efficient and attains rate of convergence . If is constant on an interval containing , the optimal rate of convergence increases to and the IIE attains this rate adaptively, i.e. without explicitly using the knowledge of local constancy. However, in this case, the asymptotic distribution is not normal. In this paper, we introduce three informed projection-type estimators of , which use knowledge on the interval of constancy and show these are all asymptotically equivalent and normal. Furthermore, we establish a local asymptotic minimax lower bound in this setting, proving that the three informed estimators are asymptotically efficient and a convolution result showing that the IIE is not efficient. We also derive the asymptotic distribution of the difference of the IIE with the efficient estimators, demonstrating that the IIE is not asymptotically equivalent to the informed estimators. Through a simulation study, we provide evidence that the performance of the IIE closely resembles that of its competitors, supporting the use of the IIE as the standard choice when no information about is available.
研究了经典Wicksell问题中平方球半径分布函数F的非参数估计。在x邻域F上的平滑条件下，Gili et al.（2024）证明了等压逆估计（IIE）是渐近有效的，其收敛速率为n/logn。如果F在包含x的区间上是常数，则最优收敛速率增加到n，并且IIE自适应地达到该速率，即不显式地使用局部常数的知识。然而，在这种情况下，渐近分布不是正态分布。本文引入了F的三个已知投影型估计，它们利用了关于常数区间的知识，证明了它们都是渐近等价的正态估计。在此基础上，我们建立了局部渐近极大极小下界，证明了这三个估计量是渐近有效的，并给出了一个卷积结果，证明了IIE是无效的。我们还推导了IIE与有效估计量之差的渐近分布，证明了IIE与知情估计量并不渐近等价。通过模拟研究，我们提供了证据，证明IIE的性能与其竞争对手非常相似，支持在没有关于F的信息时使用IIE作为标准选择。
{"title":"Asymptotically efficient estimation under local constraint in Wicksell’s problem","authors":"Francesco Gili, Geurt Jongbloed, Aad van der Vaart","doi":"10.1016/j.jspi.2025.106299","DOIUrl":"10.1016/j.jspi.2025.106299","url":null,"abstract":"<div><div>We consider nonparametric estimation of the distribution function <span><math><mi>F</mi></math></span> of squared sphere radii in the classical Wicksell problem. Under smoothness conditions on <span><math><mi>F</mi></math></span> in a neighborhood of <span><math><mi>x</mi></math></span>, in Gili et al. (2024) it is shown that the Isotonic Inverse Estimator (IIE) is asymptotically efficient and attains rate of convergence <span><math><msqrt><mrow><mi>n</mi><mo>/</mo><mo>log</mo><mi>n</mi></mrow></msqrt></math></span>. If <span><math><mi>F</mi></math></span> is constant on an interval containing <span><math><mi>x</mi></math></span>, the optimal rate of convergence increases to <span><math><msqrt><mrow><mi>n</mi></mrow></msqrt></math></span> and the IIE attains this rate adaptively, i.e. without explicitly using the knowledge of local constancy. However, in this case, the asymptotic distribution is not normal. In this paper, we introduce three <em>informed</em> projection-type estimators of <span><math><mi>F</mi></math></span>, which use knowledge on the interval of constancy and show these are all asymptotically equivalent and normal. Furthermore, we establish a local asymptotic minimax lower bound in this setting, proving that the three <em>informed</em> estimators are asymptotically efficient and a convolution result showing that the IIE is not efficient. We also derive the asymptotic distribution of the difference of the IIE with the efficient estimators, demonstrating that the IIE is <em>not</em> asymptotically equivalent to the <em>informed</em> estimators. Through a simulation study, we provide evidence that the performance of the IIE closely resembles that of its competitors, supporting the use of the IIE as the standard choice when no information about <span><math><mi>F</mi></math></span> is available.</div></div>","PeriodicalId":50039,"journal":{"name":"Journal of Statistical Planning and Inference","volume":"240 ","pages":"Article 106299"},"PeriodicalIF":0.8,"publicationDate":"2025-05-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144088647","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-05-10DOI: 10.1016/j.jspi.2025.106298
Yangbing Tang , Jiang Du , Zhongzhan Zhang
We propose a new test for the heterogeneity of the spatial autoregressive parameter in semiparametric varying-coefficient spatial autoregressive models. Our specification test is built on the difference of parametric and nonparametric estimates of the spatial autoregressive coefficient, where the two estimates are obtained by the sieve GMM estimation method. Under mild conditions, we derive the limiting null distribution, the local power property and consistency of the test statistic. Numerical simulations show promising performance of the proposed test for finite samples in the considered cases, and the crime data of Tokyo is analyzed to illustrate the usefulness of the test.
{"title":"A nonparametric test for the heterogeneity of the spatial autoregressive parameter","authors":"Yangbing Tang , Jiang Du , Zhongzhan Zhang","doi":"10.1016/j.jspi.2025.106298","DOIUrl":"10.1016/j.jspi.2025.106298","url":null,"abstract":"<div><div>We propose a new test for the heterogeneity of the spatial autoregressive parameter in semiparametric varying-coefficient spatial autoregressive models. Our specification test is built on the difference of parametric and nonparametric estimates of the spatial autoregressive coefficient, where the two estimates are obtained by the sieve GMM estimation method. Under mild conditions, we derive the limiting null distribution, the local power property and consistency of the test statistic. Numerical simulations show promising performance of the proposed test for finite samples in the considered cases, and the crime data of Tokyo is analyzed to illustrate the usefulness of the test.</div></div>","PeriodicalId":50039,"journal":{"name":"Journal of Statistical Planning and Inference","volume":"240 ","pages":"Article 106298"},"PeriodicalIF":0.8,"publicationDate":"2025-05-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143936935","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-05-07DOI: 10.1016/j.jspi.2025.106296
Huazhen Yu , Lixin Zhang
Although the independent censoring assumption is commonly used in survival analysis, it can be violated when the censoring time is related to the survival time, which often happens in many practical applications. To address this issue, we propose a flexible semiparametric method for dependent censored data. Our approach involves fitting the survival time and the censoring time with a joint transformed linear model, where the transformed function is unspecified. This allows for a very general class of models that can account for possible covariate effects, while also accommodating administrative censoring. We assume that the transformed variables have a bivariate nonnormal distribution based on parametric copulas and parametric marginals, which further enhances the flexibility of our method. We demonstrate the identifiability of the proposed model and establish the consistency and asymptotic normality of the model parameters under appropriate regularity conditions and assumptions. Furthermore, we evaluate the performance of our method through extensive simulation studies, and provide a real data example for illustration.
{"title":"Copula-based semiparametric nonnormal transformed linear model for survival data with dependent censoring","authors":"Huazhen Yu , Lixin Zhang","doi":"10.1016/j.jspi.2025.106296","DOIUrl":"10.1016/j.jspi.2025.106296","url":null,"abstract":"<div><div>Although the independent censoring assumption is commonly used in survival analysis, it can be violated when the censoring time is related to the survival time, which often happens in many practical applications. To address this issue, we propose a flexible semiparametric method for dependent censored data. Our approach involves fitting the survival time and the censoring time with a joint transformed linear model, where the transformed function is unspecified. This allows for a very general class of models that can account for possible covariate effects, while also accommodating administrative censoring. We assume that the transformed variables have a bivariate nonnormal distribution based on parametric copulas and parametric marginals, which further enhances the flexibility of our method. We demonstrate the identifiability of the proposed model and establish the consistency and asymptotic normality of the model parameters under appropriate regularity conditions and assumptions. Furthermore, we evaluate the performance of our method through extensive simulation studies, and provide a real data example for illustration.</div></div>","PeriodicalId":50039,"journal":{"name":"Journal of Statistical Planning and Inference","volume":"240 ","pages":"Article 106296"},"PeriodicalIF":0.8,"publicationDate":"2025-05-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143927463","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号