Development of C-index-based survival tree and its application to biomedical research

• Project/Area Number: 24700276
• Research Category: Grant-in-Aid for Young Scientists (B)
• Allocation Type: Multi-year Fund
• Research Field: Statistical science
• Research Institution: Osaka University
• Principal Investigator: HAYASHI Kenichi 大阪大学, 医学(系)研究科(研究院), 助教 (70617274)
• Project Period (FY): 2012-04-01 – 2015-03-31
• Project Status: Completed (Fiscal Year 2014)
• Budget Amount: ¥4,160,000 (Direct Cost: ¥3,200,000、Indirect Cost: ¥960,000); Fiscal Year 2014: ¥1,040,000 (Direct Cost: ¥800,000、Indirect Cost: ¥240,000); Fiscal Year 2013: ¥1,300,000 (Direct Cost: ¥1,000,000、Indirect Cost: ¥300,000); Fiscal Year 2012: ¥1,820,000 (Direct Cost: ¥1,400,000、Indirect Cost: ¥420,000)
• Keywords: 医学統計学 / 生存時間データ / C-index / 二重打ち切りデータ / Decision Tree / クロスバリデーション / C統計量

Outline of Final Research Achievements

This project investigated a concordance probability for time-to-event data called the C-index and its applicability to Survival Tree. The C-index can be seen as an extension of the area under the ROC curve for binary data. However, the existence of censored individuals causes a bias in estimating the C-index. We proposed some estimators for the C-index which reduce the bias, and demonstrated their superiority to the existing estimators via numerical experiments.

Research Products

• [Journal Article] Effects of unlabeled data on classification error in normal discriminant analysis (2014)
  • Author(s): Keiji Takai, Kenichi Hayashi
  • Journal Title: Journal of Statistical Planning and Inference Volume: 147 Pages: 66-83
  • DOI: 10.1016/j.jspi.2013.11.004
  • Peer Reviewed
• [Journal Article] Sample size determination in group-sequential clinical trials with two co-primary endpoints (2014)
  • Author(s): Asakura, K., Hamasaki, T., Sugimoto, T., Hayashi, K., Evans, S., Sozu T
  • Journal Title: Statistics in Medicine Volume: (First published online) Issue: 17 Pages: 2897-2913
  • DOI: 10.1002/sim.6154
  • Peer Reviewed / Open Access
• [Journal Article] Finite-sample analysis of impacts of unlabelled data and their labelling mechanisms in linear discriminant analysis (2014)
  • Author(s): Kenichi Hayashi, Keiji Takai
  • Journal Title: Communications in Statistics -Simulation and Computation Volume: - Issue: 1 Pages: 184-203
  • DOI: 10.1080/03610918.2014.957847
  • Peer Reviewed
• [Journal Article] A boosting method with asymmetric mislabeling probabilities which depend on covariates (2012)
  • Author(s): Kenichi Hayashi
  • Journal Title: Computational Statistics Volume: 27 Issue: 2 Pages: 203-218
  • DOI: 10.1007/s00180-011-0250-8
  • Peer Reviewed
• [Journal Article] Estimation of the Variance for the Maximum Likelihood Estimates in Normal Mixture Models and Normal Hidden Markov Models (2011)
  • Author(s): Masanari Iida and Naoto Niki
  • Journal Title: Journal of the Japanese Society of Computational Statistics Volume: 24 Issue: 1 Pages: 39-66
  • DOI: 10.5183
  • NAID: 10030497169
  • ISSN: 0915-2350, 1881-1337
  • Peer Reviewed / Open Access / Acknowledgement Compliant
• [Presentation] Consistent estimators of a concordance probability for doubly-censored time-to-event responses (2014)
  • Author(s): 林 賢一，清水 泰隆
  • Organizer: 2014年度統計関連学会連合大会
  • Place of Presentation: 東京大学
  • Year and Date: 2014-09-13 – 2014-09-16
• [Presentation] On the nonparametric estimation of a concordance probability for doubly-censored time-to-event response (2014)
  • Author(s): Kenichi Hayashi
  • Organizer: The 26th International Biometric Conference
  • Place of Presentation: Firenze Fiera, Italy
  • Year and Date: 2014-07-06 – 2014-07-11
• [Presentation] Simultaneous analysis of clustering and dimension reduction for binary variables with application to biomedical data (2014)
  • Author(s): Michio Yamamoto, Kenichi Hayashi
  • Organizer: The 26th International Biometric Conference
  • Place of Presentation: Firenze Fiera, Italy
  • Year and Date: 2014-07-06 – 2014-07-11
• [Presentation] 打ち切りを伴う生存時間データに対するC統計量のバイアスについて (2013)
  • Author(s): 林 賢一
  • Organizer: 日本計算機統計学会第27回大会
  • Place of Presentation: 弘前大学
• [Presentation] Evaluating discriminant performance of a semi-supervised linear discriminant analysis against a supervised one for heteroscedastic normal populations (2013)
  • Author(s): Kenichi Hayashi
  • Organizer: 2013 Joint Statistical Meeting
  • Place of Presentation: Palais des congres de Montreal, Canada
• [Presentation] Sample size evaluation in group sequential designs for clinical trials with two continuous endpoints as co-primary contrasts (2013)
  • Author(s): Koko Asakura, Kenichi Hayashi, Takashi Sugimoto, Takashi Sozu, Toshimitsu Hamasaki
  • Organizer: 2013 Joint Statistical Meeting
  • Place of Presentation: Palais des congres de Montreal, Canada
• [Presentation] Model-based clustering for multivariate binary data with dimension reduction (2013)
  • Author(s): Michio Yamamoto, Kenichi Hayashi
  • Organizer: 2013 Joint Statistical Meeting
  • Place of Presentation: Palais des congres de Montreal, Canada
• [Presentation] 多変量二値データに対する次元縮約を伴うクラスター分析法 (2013)
  • Author(s): 山本 倫生，林 賢一
  • Organizer: 2013年度統計関連学会連合大会
  • Place of Presentation: 大阪大学
• [Presentation] Clustering of multivariate binary data via penalized latent class analysis with dimension reduction (2013)
  • Author(s): Michio Yamamoto, Kenichi Hayashi
  • Organizer: The 6th International Conference of the ERCIM WG on Computational and Methodological Statistics
  • Place of Presentation: University of London, UK
• [Presentation] 生存時間データにおけるモデル評価の一考察 (2013)
  • Author(s): 林 賢一
  • Organizer: 研究集会「統計手法における近年の話題とその応用」
  • Place of Presentation: KKRホテル広島
• [Presentation] The effects of unlabeled data on a linear discriminant function for heteroscedastic normal populations
  • Author(s): 林 賢一
  • Organizer: 2012年度統計関連学会連合大会
  • Place of Presentation: 北海道大学