Does Bayesian DSGE-VAR model require all variables to be covariance statioinary? Can quarterly stock market index growth rate be used as a variable in Bayesian DSGE-VAR

Estimation DSGE-VAR
1

Dear Johannes,
First thank you very much for your previous guidance, I am grateful.
In my DSGE model, one of observed variables is quarterly stock market return, and I generated artificial data from DSGE model to formulate priors for Bayesian DSGE-VAR model,
Can I treat quarterly stock market index growth rate as covariance stationary?
I plan to use Bayesian DSGE-VAR model for forecasting, and one of the observed variables in Bayesian VAR is quarterly stock market index growth rate?, is this ok? Do you think quarterly stock market growth rate can be treated as covariance stationary for estimation of DSGE model and Bayesian DSGE-VAR model?
Does Bayesian DSGE-VAR model require all variables to be covariance statioinary?
Thank you very much and look forward to hearing from you.
Best wishes,
Jesse

2
  1. I would consider stock market growth rates as stationary. Finance people might differ in their opinion.
  2. I think the individual variables need not be stationary as long as there is cointegration, i.e. the linear combination of variables is stationary (the Sims/Stock/Watson result).