MANOVA and Discriminant Function Analysis
I am interested in performing a MANOVA but I want to build the prediction equation after running the analysis which means I need unstandardized coefficients. The MANOVA module doesn't aloow for this. The Linear Discriminant analysis in the Machine Learning module provides the linear discriminant coefficients but it is not clear whether these are standardized or unstandardized. In the analysis I ran the output included 2 linear discriminant functions (LD1 and LD2). Is there a way to know which of these accounts for more of the variability? Thanks
Comments
There are multiple issues here. First off, predictions are usually made using regression models, not factorial designs. However, there is no principled reason why this would be the case. I think you can do this in our Quality Control module under DOE, but I will check with the experts. If you want to see the prediction equation for ANOVA-style models you can post an issue on our GitHub page.
Also, "The Linear Discriminant analysis in the Machine Learning module provides the linear discriminant coefficients but it is not clear whether these are standardized or unstandardized. In the analysis I ran the output included 2 linear discriminant functions (LD1 and LD2). Is there a way to know which of these accounts for more of the variability?" -- here I will ask our experts.
EJ