What if you have taxi data from multiple states?

[gireeshkbogu]

What if you have taxi data not just from NYC but also from several different states?
You end up with redundant time stamps and probably similar or different passenger counts. How can I teach this to the model? The model should understand the variation between passenger counts from each state. Therefore the model should be able to capture the anomaly that is specific to the state. Is this possible here? Thanks.