Andrej Muhič: Hierarchical regression

The hierarchical regression is usually used to leverage the structure of the model when the data is scarce. This is especially important in environments with a very low signal to noise ratio that are highly non-stationary and there is a need to capture the local behaviour. One example of the structure would be the country specific effects. We will look at some examples and show how this models can be efficiently estimated. If time permits we will discuss what to do when the assumptions of the model are not satisfied.