CoChat
Sign Up
Cited by 809Statistical Methods in EpidemiologyArtificial Intelligence in HealthcareStatistical Methods and Inference

The DOI matches its Crossref record; 32 of the 34 references it cites were found in a registry or an index.

Modern modelling techniques are data hungry: a simulation study for predicting dichotomous endpoints

Open access
van der Ploeg, Tjeerd, Austin, Peter C, Steyerberg, Ewout W·2014·BMC Medical Research Methodology
DOI 10.1186/1471-2288-14-137PMC PMC4289553
Open this paper in CoChat — free accountOpen PDF

Abstract

BACKGROUND: Modern modelling techniques may potentially provide more accurate predictions of binary outcomes than classical techniques. We aimed to study the predictive performance of different modelling techniques in relation to the effective sample size ("data hungriness"). METHODS: We performed simulation studies based on three clinical cohorts: 1282 patients with head and neck cancer (with 46.9% 5 year survival), 1731 patients with traumatic brain injury (22.3% 6 month mortality) and 3181 patients with minor head injury (7.6% with CT scan abnormalities). We compared three relatively modern modelling techniques: support vector machines (SVM), neural nets (NN), and random forests (RF) and two classical techniques: logistic regression (LR) and classification and regression trees (CART). We created three large artificial databases with 20 fold, 10 fold and 6 fold replication of subjects, where we generated dichotomous outcomes according to different underlying models. We applied each modelling technique to increasingly larger development parts (100 repetitions). The area under the ROC-curve (AUC) indicated the performance of each model in the development part and in an independent validation part. Data hungriness was defined by plateauing of AUC and small optimism (difference between the mean apparent AUC and the mean validated AUC <0.01). RESULTS: We found that a stable AUC was reached by LR at approximately 20 to 50 events per variable, followed by CART, SVM, NN and RF models. Optimism decreased with increasing sample sizes and the same ranking of techniques. The RF, SVM and NN models showed instability and a high optimism even with >200 events per variable. CONCLUSIONS: Modern modelling techniques such as SVM, NN and RF may need over 10 times as many events per variable to achieve a stable AUC and a small optimism than classical modelling techniques such as LR. This implies that such modern techniques should only be used in medical prediction problems if very large data sets are available.

Overview

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1186/1471-2288-14-137

Cites

32 of 32 references verified0 citation errors · 0 not checkable

Statistics

Figures

Figure 1 Flow chart simulation design.
Figure 1 Flow chart simulation design.CC BY · PubMed Central
Figure 2 Validated AUC-values vs. events per variable, HNSCC cohort.
Figure 2 Validated AUC-values vs. events per variable, HNSCC cohort.CC BY · PubMed Central
Figure 3 Relative validated AUC-values vs. events per variable, HNSCC cohort.
Figure 3 Relative validated AUC-values vs. events per variable, HNSCC cohort.CC BY · PubMed Central
Figure 4 Optimism vs. events per variable, HNSCC cohort.
Figure 4 Optimism vs. events per variable, HNSCC cohort.CC BY · PubMed Central
Figure 5 Validated AUC-values vs. number of events per variable, TBI cohort.
Figure 5 Validated AUC-values vs. number of events per variable, TBI cohort.CC BY · PubMed Central
Figure 6 Relative validated AUC-values vs. events per variable, TBI cohort.
Figure 6 Relative validated AUC-values vs. events per variable, TBI cohort.CC BY · PubMed Central
Figure 7 Optimism vs. events per variable, TBI cohort.
Figure 7 Optimism vs. events per variable, TBI cohort.CC BY · PubMed Central
Figure 8 Validated AUC-values vs. events per variable, CHIP cohort.
Figure 8 Validated AUC-values vs. events per variable, CHIP cohort.CC BY · PubMed Central
Figure 9 Relative validated AUC-values vs. events per variable, CHIP cohort.
Figure 9 Relative validated AUC-values vs. events per variable, CHIP cohort.CC BY · PubMed Central
Figure 10 Optimism vs. events per variable, CHIP cohort.
Figure 10 Optimism vs. events per variable, CHIP cohort.CC BY · PubMed Central
Figure 11 Validated AUC-values vs. events per variable, CHIP5050 cohort.
Figure 11 Validated AUC-values vs. events per variable, CHIP5050 cohort.CC BY · PubMed Central
Figure 12 Relative validated AUC-values vs. events per variable, CHIP5050 cohort.
Figure 12 Relative validated AUC-values vs. events per variable, CHIP5050 cohort.CC BY · PubMed Central
Figure 13 Optimism vs. events per variable, CHIP5050 cohort.
Figure 13 Optimism vs. events per variable, CHIP5050 cohort.CC BY · PubMed Central

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Show all 34 references

Research works printed but not linked: 1

  1. Kurahashi I: KsPlot: Check the power of seveal statistical models. 2011

Cited by

Sorted by citations

Items 101–200 of 809

Related

See all

Metadata

Authors
van der Ploeg, T., Austin, P. C., & Steyerberg, E. W.
Affiliations
  • University of Toronto (CA)
  • Erasmus MC (NL)
ORCID
Venue
BMC Medical Research Methodology
Publisher
Springer Science and Business Media LLC
Published
2014-12
Volume
14
Issue
1
Pages
137-137
ISSN
1471-2288 (ISSN-L)
DOI
10.1186/1471-2288-14-137
PMC
PMC4289553
OpenAlex
W1602160603
Cited by
809
References
34
Citations by year
2026: 118 · 2025: 107 · 2024: 91 · 2023: 117 · 2022: 114 · 2021: 107 · 2020: 59 · 2019: 39 · 2018: 30 · 2017: 13 · 2016: 8 · 2015: 6
FWCI
9.51
Type
journal article
Language
en
Open access
Licence
Locations
  • BMC Medical Research Methodology (publishedVersion, cc-by)
  • Europe PMC (PubMed Central) (publishedVersion, other-oa)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • EUR Research Repository (Erasmus University Rotterdam) (publishedVersion)
  • TSpace (University of Toronto) (submittedVersion)
  • EUR Research Repository (Erasmus University Rotterdam) (submittedVersion)
Topics
  • Physical Sciences › Mathematics › Statistics and Probability › Statistical Methods in Epidemiology
  • Health Sciences › Health Professions › Health Information Management › Artificial Intelligence in Healthcare
  • Physical Sciences › Mathematics › Statistics and Probability › Statistical Methods and Inference
Keywords
effective sample size, predictive performance, binary outcome prediction, support vector machine, artificial neural networks, random forest, logistic regression, classification and regression trees, optimism, traumatic brain injury mortality, mild traumatic brain injury, CT abnormalities, model stability
MeSH terms
Brain Injuries, Data Interpretation, Statistical, Head and Neck Neoplasms, Humans, ROC Curve, Models, Statistical, Databases, Factual, Head Injuries, Closed, Treatment Outcome, Endpoint Determination, Support Vector Machine

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review