I am little confused about how to structure my specific data for multilevel analysis.

I have 10 categories and each category has some items in them.

The dataset is available for 117 weeks. There is quantity share and price index available on all of them.

My usual regression takes place with predicting the quantity mix by using the price indices for items within a category. This gives me regression coefficients for items but by treating every category separately.

I aim to use the multi level modeling approach to see the regression coefficients effects between items and categories.

This makes the model quite complicated and I’m not sure how to structure the data.


1 Answer 1


Take a look at mixed effects models. These give you a natural way to deal with heiractical and nested data.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.