Sequential LND sensitivity test for binary response data

Abstract

Sensitivity tests are used to make inferences about a sensitivity, a characteristic property of some products that cannot be observed directly. For binary response sensitivity data (dead or alive, explode or unexplode), the Langlie and Neyer are two well-known sensitivity tests. The priorities of the Langlie and Neyer tests are investigated in this paper. It is shown that the Langlie test has an advantage in getting an overlap, while the Neyer test has better estimation precision. Aiming at improving both the speed of getting an overlap and the estimation precision, we propose a new sensitivity test which replaces the first part of the Neyer test with the Langlie test. Our simulation studies indicate that the proposed test outperforms the Langlie, Neyer and Dror and Steinberg tests from the viewpoints of estimation precision and probability of obtaining an overlap.

Keywords:

• sensitivity test
• D-optimality criterion
• maximum likelihood estimate
• Fisher information matrix
• Langlie test
• Neyer test

Acknowledgements

The authors thank the Associate Editor and two anonymous referees for their valuable and constructive comments which significantly improved the manuscript. This work was supported by National Science Fund of China (11271135, 11001083, 11101156) and China Scholarship Council.