I am trying to do a multiple regression analysis based on CDS data to find out the regression coefficients and interpret them. In addition, I want to check for multicolinearity and create a forecast model based on 20% of the data. Is this workflow suitable or have I forgotten something?
1 Like
This topic was automatically closed 7 days after the last reply. New replies are no longer allowed.
Hi @Tobias_Gutzeit,
Nice workflow. Just one suggestion:
You’re using two Regression Predictor nodes, but only one Linear Regression Learner. The second predictor (for the 20% holdout set) should connect to its own learner trained on the 80% training data — otherwise, it’s reusing just the model from the last cross-validation loop.
1 Like