Your data analysis solution
Name is required.
Email address is required.
Invalid email address
Answer is required.
Exceeding max length of 5KB

logistic regression

By sandoz | Jan 21, 2016 11:02AM CET

Hi,
3 questions before I may use the free trial:
- How are categorical variables dealt with: can they be used as is or is a transformation necessary?

- Is it possible to get a probability for the dependant variable( instead of a 0 / 1 )classification)

- once a model is obtained can it be readily used on different (testing) data?
sandoz

Up

0

Down

By Efthalia | Jan 21, 2016 05:23PM CET | XLSTAT Agent

Dear Sandoz,

Please find my answers below:

1. You can use categorical variables for logistic regression without any transformation. If your dependent variables are ordered categorical then you should use the ordinal model. If these are unordered categorical variables then you should use the multinomial model.

2. The predicted probabilities are displayed in the output.

3. You can define a testing set through the dialog box in order to validate the model.

Regards,
Thalia

This question has received the maximum number of answers.

Contact Us

1c26995d494fb3061dd0ae8571ffc0a4@xlstat.desk-mail.com
https://cdn.desk.com/
false
desk
Loading
seconds ago
a minute ago
minutes ago
an hour ago
hours ago
a day ago
days ago
about
false
Invalid characters found
/customer/portal/articles/autocomplete
9283