Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Association between split selection instability and predictive error in survival trees.

M Radespiel-Tröger1, O Gefeller, T Rabenstein

  • 1Department of Medical Informatics, Biometry, and Epidemiology, Friedrich-Alexander-University,Erlangen, Germany. radespmn@imbe.imed.uni-erlangen.de

Methods of Information in Medicine
|October 5, 2006
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Correlation of factor XIII subunit A with factor XIII activity in a population of parturient women.

Archives of gynecology and obstetrics·2024
Same author

Trends and current spectrum of contact allergy in Central Europe: results of the Information Network of Departments of Dermatology (IVDK) 2007-2018.

The British journal of dermatology·2020
Same author

Impact of age on the efficacy of oxaliplatin in the preoperative chemoradiotherapy and adjuvant chemotherapy of rectal cancer: a post hoc analysis of the CAO/ARO/AIO-04 phase III trial.

Annals of oncology : official journal of the European Society for Medical Oncology·2018
Same author

Neoadjuvant rectal score as individual-level surrogate for disease-free survival in rectal cancer in the CAO/ARO/AIO-04 randomized phase III trial.

Annals of oncology : official journal of the European Society for Medical Oncology·2018
Same author

The Garment Protection Factor: further advances in labelling sun-protective clothing.

The British journal of dermatology·2018
Same author

[Small-scale analysis of suspected cancer clusters based on cancer registry data].

Bundesgesundheitsblatt, Gesundheitsforschung, Gesundheitsschutz·2017
Same journal

Design and methodological development of a digital clinical safety training programme informed by a national framework: a New Zealand case study.

Methods of information in medicine·2026
Same journal

Panic Prediction from Digital Phenotyping: Subject-Level Cross-Validation Reveals Limited Between-Person Generalization.

Methods of information in medicine·2026
Same journal

Agent-Based Modeling Approach for Population Dynamics of the Biological Vector Aedes Aegypti.

Methods of information in medicine·2026
Same journal

A Statistical Framework for Person-centered Analysis of Digital Service Use in Public Health and Social Care.

Methods of information in medicine·2026
Same journal

Assessing the Quality of Electronic Discharge Summaries: A Cross-Sectional Study Using the Validated Spanish Version of the PDQI-9.

Methods of information in medicine·2026
Same journal

A Knowledge Graph-Driven Hypergeometric Efficacy Prediction Model for Classical Traditional Chinese Herbal Formulas.

Methods of information in medicine·2026
See all related articles

Split selection instability in survival trees is linked to higher predictive error. The logrank statistic (LR) algorithm shows the lowest error, making it a recommended choice for building survival trees.

Area of Science:

  • Biostatistics
  • Machine Learning
  • Survival Analysis

Background:

  • Survival trees are valuable tools for analyzing time-to-event data.
  • Assessing the stability of split selection is crucial for reliable tree construction.
  • Predictive error is a key metric for evaluating model performance.

Purpose of the Study:

  • To investigate split selection instability in six survival tree algorithms.
  • To determine the relationship between split selection instability and predictive error.
  • To identify optimal algorithms for survival tree construction.

Main Methods:

  • Evaluated six survival tree algorithms: logrank statistic (LR), Kaplan-Meier (KM), martingale residuals (MR), Poisson regression (PR), within-node impurity (WI), and exponential log-likelihood loss (XL).

Related Experiment Videos

  • Employed a bootstrap study on a gallbladder stone patient dataset.
  • Assessed predictive error using the integrated Brier score for censored data.
  • Analyzed instability using box-percentile plots, entropy, and coefficients of variation.

    • Main Results:

    • A positive association was found between covariate selection instability and predictive error in the root node.
    • The logrank statistic (LR) algorithm demonstrated the lowest predictive error.
    • Kaplan-Meier (KM) and martingale residuals (MR) algorithms exhibited the highest predictive error.

    Conclusions:

    • Predictive error in survival trees is correlated with split selection instability.
    • The logrank statistic (LR) algorithm is recommended for constructing survival trees due to its low predictive error.
    • Unpruned survival trees with multivariate p-value adjustment can be as effective as pruned trees.