Hi, Dear All,
I am new to the PVAR model and have some questions regarding the choice of exogenous variables in the PVAR model:
1. When specifying a PVAR model, which of the following should be included in the exog() function for the exogenous control variables?
A) Variables that influence both endogenous variables (y1 and y2) or
B) Variables that influence either y1 or y2 individually
According to Chang et al. (2024) (link attached below), option (A) is recommended. However, could this approach result in omitted variable bias if I do not consider variables that influence only y1 (or only y2)?
2. A general question: When specifying a PVAR model, what types of control variables can be included in exog()? Is it correct to include variables that determine y1 and y2, rather than those that are outcomes of y1 and y2?
Any guidance on appropriate exog() variable selection for PVAR models would be helpful. Thank you.
Ref: Chang, X., Kwok, W. C., & Wong, G. (2024). Demand uncertainty, inventory, and cost structure. Contemporary Accounting Research, 41(1), 226-254.
https://onlinelibrary.wiley.com/doi/...911-3846.12918
