Studies on efficiency of sequential procedures and its applications

2016 Fiscal Year Final Research Report

• Project/Area Number: 26400193
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Research Field: Foundations of mathematics/Applied mathematics
• Research Institution: Niigata University
• Principal Investigator: Isogai Eiichi 新潟大学, 自然科学系, フェロー (40108014)
• Project Period (FY): 2014-04-01 – 2017-03-31
• Keywords: 統計数学 / 逐次解析

Outline of Final Research Achievements

1. On a two sample problem with two exponential distributions of which location and scale parameters are unknown, we considered the problem of fixed-width confidence interval estimation of a linear combination of two locations. Given a confidence coefficient and a length of interval, the minimum sample sizes include the unknown scale parameters and so we cannot use them in practice. Thus, by using the three-stage procedures, we gave the asymptotic second-order expansions of the average sample numbers and coverage probability.
2. We considered two exponential distributions with unknown location and scale parameters and investigated the bounded risk point estimation problem of a linear combination of two location parameters. With the risk less than or equal to a given upper bound, the unknown location parameters are included in the optimal sample sizes. Therefore, by using three-stage procedures we provided the asymptotic second-order expansions of the average sample sizes and the risk.

Free Research Field

数物系科学