Local Influence in Regression Models with Measurement Errors and Censored Data Considering the Student-t Distribution

被引：0

作者：

Montoya, Alejandro Monzon ^{[1
,2
]}

机构：

[1] Univ Fed Minas Gerais, Dept Estat, Ave Antonio Carlos 6-627,Campus Pampulha, BR-31270901 Belo Horizonte, MG, Brazil

[2] Univ Nacl San Cristobal Huamanga, Dept Matemat & Fis, Ave Independencia s-n,Ciudad Univ, Ayacucho 05001, Ayacucho, Peru

来源：

SANKHYA-SERIES B-APPLIED AND INTERDISCIPLINARY STATISTICS | 2024年 / 86卷 / 01期

关键词：

Censored data; ECM algorithm; measurement error models; student-t distribution; COMPARATIVE CALIBRATION; MAXIMUM-LIKELIHOOD; INCOMPLETE-DATA;

D O I：

10.1007/s13571-023-00316-6

中图分类号：

O21 [概率论与数理统计]; C8 [统计学];

学科分类号：

020208 ; 070103 ; 0714 ;

摘要：

In this paper, the local influence approach is studied in regression models with measurement errors for multivariate censored responses under the Student-t distribution. The multivariate Student-t distribution and the multivariate normal, distributions of the independent normal class, are studied and used to compare various measuring instruments. The ECM algorithm is used to obtain maximum likelihood estimates of the model parameters and using the log-likelihood function of the complete data we obtain measures of local influence based on the methodology proposed by Zhu and Lee (Journal of the Royal Statistical Society, Series B 63:121-126, 2001) and Lee and Xu (Computational Statistics and Data Analysis 45:321-341, 2004). Finally, the described methodologies are used in real data analysis that illustrates the usefulness of the approach.

引用

页码：91 / 108

页数：18

共 24 条

[21] MENG XL, 1993, BIOMETRIKA, V80, P267, DOI 10.2307/2337198
[22] Conformal normal curvature and assessment of local influence
Poon, WY
Poon, YS
[J]. JOURNAL OF THE ROYAL STATISTICAL SOCIETY SERIES B-STATISTICAL METHODOLOGY, 1999, 61 : 51 - 61
[23] R Core Team, 2023, R: A Language and Environment for Statistical Computing
[24] Local influence for incomplete-data models
Zhu, HT
Lee, SY
[J]. JOURNAL OF THE ROYAL STATISTICAL SOCIETY SERIES B-STATISTICAL METHODOLOGY, 2001, 63 : 111 - 126

← 1 2 3 →