Focus on : Contemporary Methods in Biostatistics ( I ) Regression Modeling Strategies

By E. Núñez, Ewout W. Steyerberg, Julio Núñeza. 2017

3,247 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/eb7c8f5c9c3fc90b2abccc9431ca4359c47b30cd

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Epidemiology, Patterns of Care, and Mortality for Patients With Acute Respiratory Distress Syndrome in Intensive Care Units in 50 Countries. (2016, background): “Duration of invasive mechanical ventilation, median (IQR), d 8 (4-16)”
Ordered quantile normalization: a semiparametric transformation built for the cross-validation era (2019, methodology): “methods that are trained using both the response and the covariates) must account for the tendency to overfit when making inferences [9].”
Perceived Overqualification and Collectivism Orientation: Implications for Work and Nonwork Outcomes (2020, result): “Using a small sample size may also lead to overfitting (Harrell, 2001) and limit the generalizability of findings because our results could reflect idiosyncratic sample characteristics.”
Membership Turnover and Collaboration Success in Online Communities: Explaining Rises (2011, unknown): “It is perhaps not surprising that the change in the pseudo-R 2 that we observe with the demotion model is smaller than that for the promotion model when we add the turnover variables because these statistics are sensitive to the large number of non-failure events (Harrell 2001).”
Advancing environmentally explicit structured population models of plants (2016, background): “However, this is an assumption often made in regression modelling (e.g. Harrell 2001).”
The number of subjects per variable required in linear regression analyses. (2015, background): “In a similar vein, Harrell [2] suggested that 10 SPV was the minimum required sample size for linear regression models to ensure accurate prediction in subsequent subjects.”
Single sample scoring of molecular phenotypes (2018, methodology): “The Spearman’s correlation and concordance index [22] between sample scores from the two platforms were calculated.”
An Empirical Comparison of Model Validation Techniques for Defect Prediction Models (2017, methodology): “We train defect prediction models using logistic regression [20] and measure the AUC performance.”
A standard protocol for reporting species distribution models (2020, background): “…(RMSE), log-likelihood, various variations of R 2 (Nash and Sutcliffe 1970, Nagelkerke 1991), the percentage of deviance explained (Hosmer and Lemeshow 2013) or calibration curve estimates (with the intercept quantifying bias and the slope depicting overconfidence, Harrell 2006).”
Developing prediction models for clinical use using logistic regression: an overview. (2019, methodology): “for model development, but these may have multiple issues as the data were not collected with model development or a specific clinical question in mind (1).”