Multiple regression with 2 categorical variables

[Xtina]

Hi all! I would like to know how to obtain a regression model to make predictions about the value of a dependant variable if the independent variables are categorical.I read about transforming them into Dummy variables, but I don't know how to do it in Statgraphics. I would like to estimate the Young's modulus of concrete elements from the type of dosage (D1, D2, D3 and D4) and the type of aggregates (A1, A2, A3, A4, A5), but i only have the experimental results from:

D1-A3
D1-A5
D1-A2
D2-A4
D2-A5
D2-A3
D3-A2
D4-A1
D4-A2

It is not a complete design of experiments... ¿is there any solution? Thanks in advance...

 

[GregLocock]

Of course there is /a/ solution, the question is, is it a good one?

without seeing the actual data it is hard to say what model is appropriate. You might build a simple model using say a3.d1 and a3.d2 to look at the strength of d1 to d2, and so on.



Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[IRstuff]

If nothing else, you ought to at least be able to put everything into a surface plot.

TTFN
faq731-376