logistic regression problem

[Sharnel Perera]

Hi,

 

I keep getting the following error message when trying to run a Binary Logistic Regression:

 

"Estimation failed due to numerical problem.  Possible reasons are: (1) at least one of the convergence criteria LCON, BCON is zero or too small, or (2) the value of EPS is too small (if not specified, the default value that is used may be too small for this data set)"

 

What is LCON, BCON, and EPS??

 

I am using SPSS version 15.

 

In terms of the analysis I am trying to run: the dependent variable has two levels (group A and group B), and there are 20 covariates all of which are continuous variables.  I am trying to run a stepwise (backward LR) log regression.  The data set is not large; there are less than 100 cases in total.

 

Hope you can help, thanks in advance!!!

Sharnel

 

 

==================================
Sharnel Perera
PhD Candidate
Brain Sciences Institute
400 Burwood Road,
Hawthorn, VIC 3122
AUSTRALIA
Ph:  [phone]
Fax: [phone]
==================================

[Leo Bakker-2]

Sharnel,

I can’t help you cause  you’re not telling much about your data, but about your questions the help file of SPSS you‘ll find the following answers on LCON, BCON a and EPS

Does this help?

BCON(value)	Change in parameter estimates to terminate iteration. Iteration terminates when the parameters change by less than the specified value. The default is 0.001. To eliminate this criterion, specify a value of 0.
LCON(value)	Percentage change in the log-likelihood ratio for termination of iterations. If the log-likelihood decreases by less than the specified value, iteration terminates. The default is 0, which is equivalent to not using this criterion.
EPS(value)	Epsilon value used for redundancy checking. The specified value must be less than or equal to 0.05 and greater than or equal to 10-12. The default is 10-8. Larger values make it harder for variables to pass the redundancy check—that is, they are more likely to be removed from the analysis

Greetz

Leo

 

Van: SPSSX(r) Discussion [mailto:[hidden email]] Namens Sharnel Perera
Verzonden: zondag 19 april 2009 16:39
Aan: [hidden email]
Onderwerp: logistic regression problem

 

Hi,

 

I keep getting the following error message when trying to run a Binary Logistic Regression:

 

"Estimation failed due to numerical problem.  Possible reasons are: (1) at least one of the convergence criteria LCON, BCON is zero or too small, or (2) the value of EPS is too small (if not specified, the default value that is used may be too small for this data set)"

 

What is LCON, BCON, and EPS??

 

I am using SPSS version 15.

 

In terms of the analysis I am trying to run: the dependent variable has two levels (group A and group B), and there are 20 covariates all of which are continuous variables.  I am trying to run a stepwise (backward LR) log regression.  The data set is not large; there are less than 100 cases in total.

 

Hope you can help, thanks in advance!!!

Sharnel

 

 

==================================
Sharnel Perera
PhD Candidate
Brain Sciences Institute
400 Burwood Road,
Hawthorn, VIC 3122
AUSTRALIA
Ph:  [phone]
Fax: [phone]
==================================