MétaCan
Menu
Back to cohort
Record W2119138019 · doi:10.1515/em-2012-0006

Comparison of Approaches to Weight Truncation for Marginal Structural Cox Models

2013· article· en· W2119138019 on OpenAlexafffund
Yongling Xiao, Erica E. M. Moodie, Michał Abrahamowicz

Bibliographic record

VenueEpidemiologic Methods · 2013
Typearticle
Languageen
FieldMathematics
TopicAdvanced Causal Inference Techniques
Canadian institutionsMcGill UniversityInstitut National d'Excellence en Santé et en Services Sociaux
FundersNational Institute of Allergy and Infectious DiseasesCanadian Institutes of Health ResearchJohns Hopkins Bloomberg School of Public HealthFeinberg School of MedicineUniversity of California, Los AngelesDeutsches KrebsforschungszentrumNatural Sciences and Engineering Research Council of CanadaUniversity of PittsburghJohns Hopkins UniversityNational Cancer InstituteFonds Québécois de la Recherche sur la Nature et les TechnologiesMcGill UniversityNorthwestern University
KeywordsTruncation (statistics)EstimatorWeightingMarginal structural modelMean squared errorMathematicsStatisticsPercentileApplied mathematicsComputer scienceEconometricsMathematical optimizationConfounding

Abstract

fetched live from OpenAlex

Marginal structural Cox Models (Cox MSMs) have been used to estimate the causal effect of a time-varying treatment on the hazard when there exist time-dependent confounders, which are themselves also affected by previous treatment. A Cox MSM can be estimated via the inverse-probability-of-treatment weighting (IPTW) estimator. However, IPTW estimators suffer from large variability if some observations are assigned extremely high weights. Weight truncation has been proposed as one simple solution to this problem, but truncation levels are typically chosen based on ad hoc criteria that have not been systematically evaluated. Bembom et al. proposed data-adaptive selection of the optimal truncation level using the estimated mean-squared error (MSE) of a truncated IPTW estimator for cross-sectional data. Based on a similar principle, we proposed data-adaptive approaches to select the truncation level that minimizes the expected MSE for time-to-event data with time-varying treatments. The expected MSE is approximated by using either observed statistics as a proxy for the true unknown parameter or using cross-validation. Simulations confirm that simple weight truncation at high percentiles such as the 99th or 99.5th of the distribution of weights improves the IPTW estimators in most scenarios we considered. Our newly proposed approaches exhibit similarly good performance and may be applied in a wide range of settings.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.

How this classification was reachedexpand

Full frame machine prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.

metaresearch head score (Codex)0.060
metaresearch head score (Gemma)0.149
Version: metacan-v3-hybrid-931329e0061cValidation status: machine_predicted_unvalidated
Candidate categoriesMetaresearch
Consensus categoriesnone
DomainCandidate signal: Methods · Consensus signal: none
Study designCandidate signal: Simulation or modeling · Consensus signal: Simulation or modeling
GenreCandidate signal: Methods · Consensus signal: Methods
Teacher disagreement score0.940
Threshold uncertainty score0.315

Distilled classifier scores by category (both heads)

CategoryCodexGemma
Metaresearch0.0600.149
Meta-epidemiology (narrow)0.0010.001
Meta-epidemiology (broad)0.0020.003
Bibliometrics0.0020.002
Science and technology studies0.0010.002
Scholarly communication0.0020.004
Open science0.0050.005
Research integrity0.0020.004
Insufficient payload (model declined to judge)0.0030.001

Machine scores (provisional)

The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.

Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.

Opus teacher head0.843
GPT teacher head0.600
Teacher spread0.243 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it

Classification

machine, unvalidated

Machine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.

Study designSimulation or modeling
DomainMethods
GenreMethods

How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".

Quick stats

Citations98
Published2013
Admission routes2
Has abstractyes

Explore more

Same venueEpidemiologic MethodsSame topicAdvanced Causal Inference TechniquesFrench-language works237,207