____?????? I would like to know whether we can calculate C-statistic using SPSS >> 13.
> If by the "c-statisitc" you are referring to the measure of the > discriminative power of the logistic equation, you can calculate it by > saving the predictive probabilities from the logistic regression > analysis and running a ROC curve with the preditive probabilities as > the "test variable." The c-statistic is the area under the curve > value.
HTH
In R/S-Plus you can just use the lrm function in the Design package or:
mean.rank <- mean(rank(x)[y == 1]) c.index <- (mean.rank - (n1 + 1)/2)/(n - n1)
(where n1 is the number of observations with y=1)
or use the somers2 function in the Hmisc package.
(where n1 is the number of observations with y=1)
or use the somers2 function in the Hmisc package.
Frank Harrell
Žiadne komentáre:
Zverejnenie komentára