Using Classification Error Costs in Building Logistic Regression Models

[JonDickens1607]

When you use a Binary Logistic Regression Model to estimate the probability of an event you can then set a threshold value

for this probability so that if P > K then classify the case as Event = 1 otherwise classify the case as Event = 0

There are four possible decisions :

Classify the case as Event = 0  when it is in fact Event = 0

Classify the case as Event = 0  when it is in fact Event = 1     TYPE 2 ERROR  [ Cost of False Negative Error = X ]

Classify the case as Event = 1  when it is in fact Event = 0     TYPE 1 ERROR  [ Cost of False Positive   Error = Y ]

Classify the case as Event = 1  when it is in fact Event = 1

Question:

If I can estimate the values of the Costs of making either a Type 1 Error or  a Type 2 Error before I start to build the Logistic Regression Model, how can I include this information in the Proc Logistic Code in order to optimise my model with respect to

 the Error Costs.

Without this inclusion, the default is to assume that the costs are all the same, which in my project is clearly not the case. 

I would appreciate it if you could provide guidance w.r.t. SAS Procedures, Documentation , SAS Code as well as Published Papers / Case Studies.

Thanks 

[Ksharp]

It looks like you want do some Bayes analysis.

Check PEVENT= option of MODEL in PROC LOGISTIC.

or you could check PROC MCMC .