Programming the statistical procedures from SAS

Conditional Logistic Regression, Approximation Method

Accepted Solution Solved
Reply
Occasional Contributor
Posts: 6
Accepted Solution

Conditional Logistic Regression, Approximation Method

 Dear Community,

 

I am wondering if SAS has similar approximation procedure when fits the conditional logistic regression as R does?

 

I.e. In R, there is an option to specify likelihood approximation method as "efron" or "breslow".

 

I don't think PROC LOGISTIC has this option.

 

Thank you!

 

 


Accepted Solutions
Solution
‎01-18-2017 10:10 AM
Super Contributor
Posts: 271

Re: Conditional Logistic Regression, Approximation Method

Sorry, I made a typing error. dummytime should be 1 for cases and 2 for controls (not 0). And the dummytime should be censored, so no riskset after 1:
model dummytime*dummytime(2)/ties=discrete

View solution in original post


All Replies
Grand Advisor
Posts: 16,842

Re: Conditional Logistic Regression, Approximation Method

There are several proc that perform logistic regression, PROC GLM and PHREG are others.

This may be what you're looking for:
http://support.sas.com/documentation/cdl/en/statug/68162/HTML/default/viewer.htm#statug_phreg_exampl...

Super Contributor
Posts: 271

Re: Conditional Logistic Regression, Approximation Method

It is true that proc logistic can not make the Breslow or Efron approximation. But, construction af variable "dummytime" with values 1 for cases and 0 for controls, and analyze this as survival analysis is equivalent to conditional logistic regression when using the option ties=discrete in PROC PHREG. If, instead, the option ties=breslow (default) or ties=efron is used, then you have the approximation.

Solution
‎01-18-2017 10:10 AM
Super Contributor
Posts: 271

Re: Conditional Logistic Regression, Approximation Method

Sorry, I made a typing error. dummytime should be 1 for cases and 2 for controls (not 0). And the dummytime should be censored, so no riskset after 1:
model dummytime*dummytime(2)/ties=discrete
Occasional Contributor
Posts: 6

Re: Conditional Logistic Regression, Approximation Method

I assume it should be the same as 

model dummytime*Y(0)/ties=discrete

given that Y is binary with value 1 and 0?

 

 

Super Contributor
Posts: 271

Re: Conditional Logistic Regression, Approximation Method

Yes, that wil work also. If those with dummytime=2 is the same as those with Y=0, then

model dummytime*Y(0)/ties=discrete

is equivalent to

model dummytime*dummytime(2)/ties=discrete

 

given of course that the rest has dummytime=1.

☑ This topic is SOLVED.

Need further help from the community? Please ask a new question.

Discussion stats
  • 5 replies
  • 152 views
  • 6 likes
  • 3 in conversation