Cointegrated x and y in time series analysis

by Simon de Fauconval   Last Updated October 09, 2019 21:19 PM

When trying to forecast with a linear model a variable using variables that are cointegrated with the variable you want to forecast, how would you expect the covariance of the error terms to be? Do you expect high covariance of the error terms?

Would using a vector autoregression model overcome any problem?

I am approaching now this new field and any input is more than welcome.

Thank you

Related Questions

Cointegration vs regression of differences

Updated October 17, 2018 12:19 PM

Vector Autoregression Model

Updated October 10, 2019 09:19 AM