Clever Geek Handbook

"Unexpected lpSolveAPI error result for primary test" for logistic regression

I am trying to create a logistic regression model in R and I am also using the safeBinaryRegression package. When I enter my model code and hit enter, I get the following message:

Delimiter error (X, Y, target = "find"): unexpected result from lpSolveAPI for primary test

I'm not sure what this error means and how to resolve it. Does this mean that there is something wrong with my data or something else?

I am very grateful if someone can help me understand this error and how to solve it.

Great value your time and ideas in advance.

+3


source to share


2 answers


I had the same problem. It eventually went away when I converted the answer from a numeric variable to a coefficient.



+1


source


Does this mean there is something wrong with my data?

Potentially, yes. This error message is the target of the package safeBinaryRegression

. You must modify your code to include separation = 'find'

to identify the terms that cause maximum likelihood estimation problems:



library(safeBinaryRegression)
mod <- glm(y ~ x, family = binomial, separation = 'find')
0


source






More articles:


All Articles