| Project/Area Number |
15K00057
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Multi-year Fund |
|---|
| Section | 一般 |
|---|
| Research Field |
Statistical science
|
|---|
| Research Institution | Kindai University |
|---|
Principal Investigator |
CHIBA Yasutaka 近畿大学, 医学部附属病院, 准教授 (80362474)
|
|---|
| Project Period (FY) |
2015-04-01 – 2019-03-31
|
| Project Status |
Completed (Fiscal Year 2018)
|
| Budget Amount *help |
¥4,160,000 (Direct Cost: ¥3,200,000、Indirect Cost: ¥960,000)
Fiscal Year 2018: ¥910,000 (Direct Cost: ¥700,000、Indirect Cost: ¥210,000)
Fiscal Year 2017: ¥1,170,000 (Direct Cost: ¥900,000、Indirect Cost: ¥270,000)
Fiscal Year 2016: ¥910,000 (Direct Cost: ¥700,000、Indirect Cost: ¥210,000)
Fiscal Year 2015: ¥1,170,000 (Direct Cost: ¥900,000、Indirect Cost: ¥270,000)
|
|---|
| Keywords | 因果推論 / 統計的仮説検定 / ランダム化試験 |
|---|
| Outline of Final Research Achievements |
In randomized trials in which two treatment groups are compared with a binary outcome, exact tests are statistical hypothesis tests for the sharp causal null hypothesis that there is no treatment effect for all individuals. The rejection of sharp causal null hypothesis does not mean that the weak causal null hypothesis, which is a hypothesis that causal risks are equal between two treatment groups, is rejected. Therefore, we cannot prove that the average treatment effect is not zero by the sharp causal null hypothesis.
In this research, I discussed methods of statistical hypothesis tests for the weak causal null hypothesis. I proposed such a method as a natural extension of Fisher’s exact test. Furthermore, I extended the method to several situations such as stratified analysis, the case of an ordinal outcome, and Bayesian approach.
|
| Academic Significance and Societal Importance of the Research Achievements |
本研究で提示した方法を用いることにより、ランダム化試験においてweakな因果帰無仮説に対する統計的仮説検定を、正確な方法として行うことができる。また、この検定方法に対応する信頼区間も構成することができる。これらにより、平均因果効果の推測をより正確に行うことができる。
将来的には、薬剤評価のための臨床試験などに、本研究で提示した方法あるいは類似の改良型が用いられることになる可能性がある。そうなれば、臨床試験における薬剤評価の方法が変わることになる。
|
