Fitting regression models with response-biased samples
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
This paper extends the work in Lawless, Kalbfleisch, & Wild (1999) on fitting regression models with response-biased samples, that is, samples where some or all the covariates are missing for some units and the probability that this happens depends in part on the value of the reponse of that unit. In general, the resulting likelihood depends on the distribution of the covariates but we are only interested in methods that do not involve modelling this distribution. We look at a variety of methods based on estimating equations, at the relationship of these methods to semi-parametric efficient methods in cases where such methods exist, and show ways of obtaining efficiency gains that can sometimes be dramatic. The Canadian Journal of Statistics 39: 519–536; 2011 © 2011 Statistical Society of Canada Cet article generalise les travaux de Lawless, Kalbfleisch et Wild (1999) sur l'ajustement de modeles de regression pour des echantillons avec biais du a la reponse, c'est-a-dire des echantillons pour lesquels quelques ou toutes les covariables sont manquantes pour quelques unites et la probabilite que cela se produise depend de la valeur de la variable reponse de ces unites. En general, la vraisemblance resultante depend de la distribution des covariables, mais nous sommes uniquement interesses aux methodes qui n'impliquent pas la modelisation de cette distribution. Nous considerons une variete de methodes basees sur les equations d'estimation et a la relation entre ces methodes et les methodes semi-parametriques efficaces lorsque celles-ci existent. Nous montrons des facons d'obtenir des gains d'efficacite qui peuvent parfois etre tres importants. La revue canadienne de statistique 39:519–536;2011 © 2011 Societe statistique du Canada
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.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.002 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
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.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it