1
$\begingroup$

Let’s say we will investigate if there is a relationship between the minimum wage and unemployment. For this, we will use the data that belong to all the states over last four years.

Should we take the average over all years for each state as the four years of data belonging to a single state may not be independent of each other? Or is it okay to keep each (year-state) data as an individual data point?

Similarly, let’s say that we will investigate if there is a relationship between the service level and sales. For this, a store chair has data for 20 stores for four weeks. Should we use the average values to represent each store? Or is it okay to use the weekly data and therefore to have four data points that correspond to each store?

$\endgroup$
1
$\begingroup$

Your datasets are examples of what is known as panel data:

http://en.wikipedia.org/wiki/Panel_data

You might have to assume something about temporal and/or cross-sectional dependency structure if you want to use panel data. There are some tests thought which can be used to test for example whether some covariance assumption is valid or not.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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