Print  Close

Robustness of Best Linear Unbiased Estimators based on Order Statistics

Masuma Mannan Co-Author
LSUHSC School of Public Health Biostatistics and Data Science Program
 
Evrim Oral Co-Author
LSUHSC School of Public Health
 
Nubaira Rizvi First Author
 
Nubaira Rizvi Presenting Author
 
Monday, Aug 5: 9:50 AM - 10:05 AM
3615 
Contributed Papers 
Oregon Convention Center 
Recently, Sanaullah et al (2019) and Ahmad et al (2023) utilized a Best Linear Unbiased Estimator (BLUE) that is based on order statistics to propose novel ratio-type estimators in survey sampling. While they studied the properties of the proposed ratio estimators in the survey sampling setting, the robustness properties of the BLUE-type estimators they used have not been thoroughly investigated. Therefore, in this study, we evaluate the robustness properties of the BLUE-type location estimators and compare them to other well-known robust estimators such as Huber's M and Tiku's modified maximum likelihood estimator using an extensive simulation study. Additionally, we demonstrate the performance of the estimators through a real-life example.

Keywords

BLUE

Modified Maximum Likelihood

Order Statistics

Location-scale families


Robustness 

Main Sponsor

IMS