Hi,
this is my code:
model<-glm(Risk ~ Status + Duration + Credit_History + Credit_amount +
Savings +
Employment + Inst_rate + Residence + Property + Age + plans +
Housing + Credits + Job + Liable + Telephone + Foreign_worker +
Purpose_A40 + Purpose_A41 + Purpose_A42 + Purpose_A43 +
Purpose_A44 + Purpose_A45 + Purpose_A46 + Purpose_A48 + Purpose_A49 +
P_Status_A91 + P_Status_A92 + P_Status_A93 + P_Status_A94 +
Others_A101 + Others_A102 + Others_A103,data=Training_Set,
family=binomial(logit))
########################### Plotting the ROC curve ###########################
ROC<-roc.plot(model$fitted.values,Training_Set["Risk"],show.thres=FALSE,main="My
ROC Plot",xlab="1-specifity",ylab="sensitivity")
the glm works fine, but I need to do further analysis on it (roc.plot etc.), so
I wrote the last line (from above), and got this error:
"Error: subscript out of bounds"
The original dataset is 800 by 34, and the glm was conducted on a training set
of the 600 first rows.
Can someone help?
---------------------
Have a wonderful day.
David Einhorn, Israel.
972+525058538
einhorn01@gmail.com
[[alternative HTML version deleted]]