Likelihood ratio tests for testing for multiple contaminants in the shocks and labelled slippage models

Abstract

In the literature related to the study of lifelengths of experimental units, little attention has been paid to the models where shocks to the units generate outliers. In the present article, we consider a situation where n experimental units under investigation receive shocks at several time points. The parameter values of the lifelength distribution may change due to each shock, resulting in the generation of outliers. We derive the likelihood ratio test statistic to investigate if the shocks have significantly altered the parameter values. We also derive a likelihood ratio test under the labelled slippage alternative with multiple contaminations. Monte Carlo studies have been carried out to investigate the power of the proposed test statistics.

Keywords:

• exponential distribution
• shocks model
• labelled slippage model
• likelihood ratio test

Acknowledgements

We thank the referee for suggestions and comments which improved this article. This research was partially supported by the Nanyang Technological University AcRF grant, Singapore and NSERC grant, Canada.