2014
DOI: 10.1002/sim.6100
|View full text |Cite
|
Sign up to set email alerts
|

A new calibration test and a reappraisal of the calibration belt for the assessment of prediction models based on dichotomous outcomes

Abstract: Calibration is one of the main properties that must be accomplished by any predictive model. Overcoming the limitations of many approaches developed so far, a study has recently proposed the calibration belt as a graphical tool to identify ranges of probability where a model based on dichotomous outcomes miscalibrates. In this new approach, the relation between the logits of the probability predicted by a model and of the event rates observed in a sample is represented by a polynomial function, whose coefficie… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
101
0

Year Published

2014
2014
2021
2021

Publication Types

Select...
7
1

Relationship

0
8

Authors

Journals

citations
Cited by 118 publications
(101 citation statements)
references
References 15 publications
0
101
0
Order By: Relevance
“…The H-L calibration was plotted using the PredictABEL library and the calibration belt was plotted using the GiViTI calibration belt library [30,34,35]. …”
Section: Methodsmentioning
confidence: 99%
“…The H-L calibration was plotted using the PredictABEL library and the calibration belt was plotted using the GiViTI calibration belt library [30,34,35]. …”
Section: Methodsmentioning
confidence: 99%
“…In the present study, the GiViTI calibration belt was used. Even though the GiViTI belt and H‐L test have been found to generate similar results, the different methods of exploring model performance might have contributed to the different results observed.…”
Section: Discussionmentioning
confidence: 96%
“…To assess calibration, the GiViTI (Gruppo Italiano per la Valutazione degli Interventi in Terapia Intensiva) calibration belt was used, with its R package (©G. Nattino and S. Finazzi). The GiViTI test was specifically designed to demonstrate visually the relationship between observed and predicted outcomes by fitting a polynomial function between the two, and calculating the 80 and 95 per cent confidence intervals.…”
Section: Methodsmentioning
confidence: 99%
“…However, when the H-L test indicates poor calibration (p<0.05) one may utilize the GiViTI calibration belt to assess the signifi cance of miscalibration and pinpoint risk intervals of either under-or overprediction. 126,211,241 Future studies are necessary to gain a better understanding of the benefi ts and possible pitfalls of using the GiViTI calibration belt.…”
Section: Sta S Cal Considera Onsmentioning
confidence: 99%