# How to find effects of several mostly categorial variables on a numeric value?

I have the following problem:

I want to analyze the following data:

Sales of products per year with respect to

• product type
• product group (similar products types are grouped together)
• country where it was sold
• kind of application product was sold for

So except the year all of those variables are categorial.

What statistical methods could be used to create a prognose for the future and to analyze the existing data and the effects and interactions of the above mentioned factors on the sales figures?

If it would be numerical factors, I'd think a regression could help - what can be done in this case? (My knowledge in statistics is quite limited up to now.)

-
Regression can handle categorical predictors with no problem. Most statistical packages (e.g. R, SAS) will dummy code the categorical variables for you and allow various types of parameterization (effect coding, reference coding, etc) – Peter Flom Oct 14 '12 at 21:59
Hi Peter, thank you very much! Please make an answer out of your comment, so I could accept it. Maybe you could add some info which tests would make sense to use to find the quality of the data.. – Martin Oct 15 '12 at 2:02