Pairwise Collinearity Pattern.pdf (102.7 KB)

Dear Johannes,

First thank you very much for your previous help, I am grateful.

i have obtained pairwise collinearity pattern in identification/sensitivity analysis, please refer to lower graph (Collinearity patterns with 1 parameter(s)) in the PDF attachment,

In my DSGE model, the general structure of a shock is as follows:

lnU=(1-rhoU)*Steady_State_U+rhoU*lnU(-1)+e

e~N(0,sigma^2)

I find likelihood’s sensitivity to rhoU and likelihood’s sensitivity to sigma are highly correlated according to the graph (yellow spots)

the general measurement equation is as follows:

observed variable=constant+ln(state_variable/steady_state_variable)

I find likelihood’s sensitivity to observed production’s constant is highly correlated with likelihood’s sensitivity to other observables’ constants according to the graph (yellow spots)

**Could you please have a look at the graph, do you think weak identification problem is serious according to my collinearity graph?**

my model is identified according to the following output (also refer to upper graph in PDF attachment)

==== Identification analysis ====

Testing prior mean

All parameters are identified in the model (rank of H).

All parameters are identified by J moments (rank of J)

Monte Carlo Testing

Testing MC sample

All parameters are identified in the model (rank of H).

All parameters are identified by J moments (rank of J)

==== Identification analysis completed ====

Thank you very much and look forward to hearing from you.

Best wishes,

Jesse