Estimate VAR; Intervention
Posted: Thu Feb 22, 2018 8:55 am
Hello, I would like to estimate a VAR with two endogenous variables: X1 & X2. X1 is present in the sample; X2 is added later in the sample. I'd like to see if presence of X2 affects X1.
I considered an intervention (dummy) variable to be treated as exogenous: 0 for when x2 is not present, 1 if present.
Is there any theoretical reason this is not appropriate? Any papers you could advise me on reading?
Thanks in advance for any help!
I considered an intervention (dummy) variable to be treated as exogenous: 0 for when x2 is not present, 1 if present.
Is there any theoretical reason this is not appropriate? Any papers you could advise me on reading?
Thanks in advance for any help!