Confidence interval construction in massive data sets

Abstract

This paper treats the problem of constructing confidence interval with massive data. Two methods using the divide and conquer strategy are proposed. The corresponding hypothesis testing problem is also developed. Theoretical and numerical results are presented to illustrate their effectiveness.

Keywords:

• Confidence interval
• coverage and length
• divide and conquer
• hypothesis testing
• massive data
• optimal local power

Acknowledgements

We would like to thank the editor and anonymous referees for their constructive comments and suggestions which have considerably improved this paper.