[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
st: Calibration curves after logistic regression
Ikke sensitiv - please ignore this message - due to in-house security
The calibration of a prognostic score may be examined by Hosmer-Lemeshow goodness-of-fit statistics. Another way to examine (and present it) is by so called calibration curves, i.e. a plot of observed f.ex. mortality (with 95% CI) against predicted mortality within deciles of risk with a diagonal representing perfect predictive ability.
Does anyone have a suggestion how to create such calibration curves after a logistic regression model in Stata 8 ?
Thank you in advance.
Per Ivar Kaaresen
University Hospital Northern Norway
* For searches and help try: