How to define quadratic model without SO() or PQ() functions ?

Machine Learning and Modeling
1

Dear all
I did not find the coding rules to express quadratic formula in Anova. The SO() & PQ() command allow to build such a model, but unfortunately, the resulting info is not separated along terms of the model but grouped instead. This command
Quad_SO <- aov(Data ~ SO(H1,H2,H3))
followed by : anova(Quad_SO)" gives :

                                          Df   Sum Sq Mean      F value     Pr(&gt;F)

SO(H1, H2, H3, H4, H5) 20 9.7 0.48 125.33 2.2e-16

Even worst,
rsm_SO <- rsm(Data~ SO(H1,H2,H3))
gives an horrible terms naming which makes sensitivity bargraph unreadable...
..... **TWI(H1, H2, H3)H1:H3...... PQ(H1, H2, H3)H3^2

I have not found any relevant example on how generate quadratic model using symbolic notation or a way to expand terms from SO or PQ commands
Thanks a lot for you help...

2

Hi!

To help us help you, could you please prepare a reproducible example (reprex) illustrating your issue? Please have a look at this guide, to see how to create one:

3

I never succeeded in using this "facility"...
I will rebuild my question just as a generic and simpler question...
Thanks anyway

4

Hi, have you tried following Andrés' guide? I think it's very clear. Otherwise, you could try this other one, which is probably as simple as it can get :slight_smile:

just note that now you don't need to use the dev version anymore, you can directly use the CRAN one. Give these guides a try! Learning to use reprexes is fundamental in order to get the most from this community.

1 Like
5

This topic was automatically closed 21 days after the last reply. New replies are no longer allowed.

If you have a query related to it or one of the replies, start a new topic and refer back with a link.