Multiple cases deletion measures in linear measurement error models

ABSTRACT

In this paper, we define a multiple cases deletion model (MCDM) in linear measurement error models (LMEMs). Then, by using the corrected score method of Nakamura (1990), the estimation of parameters is obtained. Furthermore, Based on MCDM, we provide computationally inexpensive deletion diagnostic tools for LMEMs. An example illustrates that our method is useful for diagnosing influential subsets of observations.

KEYWORDS:

• Corrected score method
• Influential observations
• LMEM
• multiple cases deletion.

Acknowledgments

I wish to thank the referee for giving valuable comments which enhanced the presentation of this paper.