Evaluation of the Performance of Survival Analysis Models: Discrimination and Calibration Measures

By Ralph B. D’Agostino, B. Nam. Advances in Survival Analysis, 2003

393 citations recorded by the source corpus.

Subfields: Economics and Econometrics, Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/f503397d6a65edbd6c82ec3f4a132a8f7b4e9bee

How researchers see this paper

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

In Economics and Econometrics

In Statistics and Probability

Citations in the corpus

Evaluating the added predictive ability of a new marker: From area under the ROC curve to reclassification and beyond (2008, methodology): “When evaluating the performance of a model after addition of a new marker, it is essential to check for improvement (or at least no adverse effect if other measures improve) in calibration, which can be quantified by, for example, the Hosmer‐Lemeshow’s chisquare or its modifications [ 32 , 33].”
Risk prediction models: I. Development, internal validation, and assessing the incremental value of a new (bio)marker (2012, methodology): “The Hosmer and Lemeshow test was computed by comparing the estimated probability of CVD from the model to the observed probability of CVD from KaplaneMeier estimates within subgroups of participants ranked by increasing estimated probabilities.(69) Table 1 shows the final ADVANCE model.”
Clinical utility of different lipid measures for prediction of coronary heart disease in men and women. (2007, methodology): “Calibration evaluates the degree of correspondence between the predicted probability of CHD based on a model and the observed CHD incidence and is typically evaluated with the modified Hosmer-Lemeshow 2 statistic.(18,29) Small values indicate a good calibration while values exceeding 20 indicate sig”
A predictive model for progression of chronic kidney disease to kidney failure. (2011, methodology): “Wecomparedtheobservedvspredictedriskofkidneyfailure foreachquintile of predicted risk and determined the magnitude of the deviation using the Nam and D’Agostino 2 statistic.(24) Goodness of Fit.”