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

Regression with qualitative/categorical explanatory variables

By Jon | Mar 04, 2018 08:01PM CET

I have an analysis with 3 Categorical/qualitative explanatory variables (with maximum 20 categories and a numerical dependent variable. Do i need to create dummy variables before I run the linear regression in XLSTAT? Does it do this for me as part of the programme?

Thanks

Up

0

Down

By Efthalia | Mar 05, 2018 11:51AM CET | XLSTAT Agent

Hello Jon,

You can run an ANCOVA analysis in this case. No need to transform your variables. Here is an example:
https://help.xlstat.com/customer/en/portal/articles/2062234-ancova-analysis-in-excel-tutorial?b_id=9283

Regards,

Up

0
By cristina casetta | Nov 19, 2018 11:52AM CET

In the chart of PLS-DA (observations), validation and prediction samples (based on the iris demo data) gives a not correctly representation of the prediction and validated samples compared to the training set and to the results of the confusion matrix.
I'm expected that when correctly classify validation and prediction samples belong in the corrected classes. This is not verify. Have you suggestions on how to interpret this chart?
Thank you in advance

Up

0

Down

By Efthalia | Nov 20, 2018 05:29PM CET | XLSTAT Agent

Hello Cristina,

Could you please send us at support@xlstat.com your XLSTAT output file so that we can have a look at your chart?
Thank you in advance.

Regards,

Post Your Public Answer

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