On the net there are tons of examples of recurrent neural network usage to "learn" sequences like music, texts, or other kind of ordered patterns, but in economics often such sequences are interdependent, in the sense that the outcome at order (time) t of one specific sequence depends also on the present and past outcomes and on the explanatory variables of other interdependent sequences (e.g. the GDP of a region depends also to the GDP, employment, etc on other regions, or COVID contagions in one region depend on those in other regions...).

Which are the approaches to "learn" these kind of sequences? My main concern is the explosion of parameters that a naive approach could generate...



Your Answer

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