| Project/Area Number |
16K16014
|
|---|
| Research Category |
Grant-in-Aid for Young Scientists (B)
|
| Allocation Type | Multi-year Fund |
|---|
| Research Field |
Statistical science
|
|---|
| Research Institution | The Institute of Statistical Mathematics (2018)
Chiba University (2016-2017) |
|---|
Principal Investigator |
Nagashima Kengo 統計数理研究所, 医療健康データ科学研究センター, 特任准教授 (20510712)
|
|---|
| Project Period (FY) |
2016-04-01 – 2019-03-31
|
| Project Status |
Completed (Fiscal Year 2018)
|
| Budget Amount *help |
¥2,600,000 (Direct Cost: ¥2,000,000、Indirect Cost: ¥600,000)
Fiscal Year 2018: ¥910,000 (Direct Cost: ¥700,000、Indirect Cost: ¥210,000)
Fiscal Year 2017: ¥1,040,000 (Direct Cost: ¥800,000、Indirect Cost: ¥240,000)
Fiscal Year 2016: ¥650,000 (Direct Cost: ¥500,000、Indirect Cost: ¥150,000)
|
|---|
| Keywords | 生存時間解析 / 情報量規準 / サンプルサイズ設計 / モデル選択規準 / レアイベントデータ / 情報量基準 / モデル選択基準 / 統計数学 / サンプルサイズ |
|---|
| Outline of Final Research Achievements |
Information criteria for Firth's penalized partial likelihood approach in Cox regression models were proposed and theoretical justification of the Heinze and Schemper's estimator was provided (Nagashima & Sato, Statistics in Medicine 2017, DOI: 10.1002/sim.7368). An AIC-type information criterion based on the risk function and BIC-type criterion were evaluated, which performed well under rare event data. The proposed AIC-type criterion was applied to prospective observational study data. It proved that the conditions of bias reduction for the cumulants in the Heinze & Schemper's estimator holds.
Moreover, sample size calculations for the Kaplan-Meier estimator in single-arm survival studies were proposed, and it was shown that the existing method uses an inappropriate approximation that can influence the accuracy especially under rare event data. This result was submitted to a peer-review journal.
|
| Academic Significance and Societal Importance of the Research Achievements |
本研究では，特にレアイベントデータにおけるCox回帰モデルにおける，各種統計量の漸近的性質等を明らかにし，新たな手法を提案した．これらの結果は同様の条件下でのさらなる応用・手法開発に繋がると考えられ，本研究の学術的意義は高いと考えられる．また，実際の臨床研究データへの適用や統計ソフトウェアの公開など，手法を実際に利用する際に活用できる参考情報も提供した．将来的には，臨床医学分野での共同研究を通じ，既存の方法では妥当な評価ができないケースにおける応用を目指す．臨床研究の成果として新たな知見が得られれば，実社会に貢献することが可能であり，意義深い結果をもたらすことが期待される．
|
