SW 2007 model

[ekaya]

In SW (2007) model, Smets and wouters didn't multiply with 100 enflation and interest rate variables and the others multiplied with 100.
Why
Thanks

[jpfeifer]

According to their data Appendix, the inflation rate is multiplied by 100 and the interest rate is already measured in percentage points (i.e. the raw series is already multiplied by 100)

[ekaya]

I will ask a question too.
Smets and Wouters (2007) didn't take first differenced interest rate variable, I wonder this why it is.
thanks again
you solved my problem every time.

[jpfeifer]

Because the interest rate is already stationary and does not need to be detrended. In contrast, output has a trend.

[ekaya]

Thank you very much for helping me solve this problem.
and what about confidence interval?
How can we interpret it?Is it important?
and tha other question ı wonder, why Smets and wouters multiplied variables with 100
thanks again

[ekaya]

Thank you very much for helping me solve this problem.
and what about confidence interval?
How can we interpret it?Is it important?
and tha other question ı wonder, why Smets and wouters multiplied variables with 100
thanks again

[jpfeifer]

See http://www.dynare.org/phpBB3/viewtopic.php?f=1&t=5697

There are no confidence intervals in SW 2007. What do you mean?

[ekaya]

Mr. Johannes Pfeifer,

I am grateful to you

like this %5 %95, what does it mean?, Is it important?
How can we interpret it?

Prior distribution Posterior distribution
Distr. Mean St.Dev. Mode Mean 5% 95%
ϕ Normal 4.00 1.50 5.48 5.74 3.97 7.42
σ c Normal 1.50 0.37 1.39 1.38 1.16 1.59
h Beta 0.70 0.10 0.71 0.71 0.64 0.78
ξ w Beta 0.50 0.10 0.73 0.70 0.60 0.81
σ L
Normal 2.00 0.75 1.92 1.83 0.91 2.78
thanks again

[jpfeifer]

Yes, it is important. It refers to the prior and posterior coming from Bayesian estimation. If you are not familiar with these concepts, you should not be estimating a model using these techniques. Before proceeding, take a look at e.g. An/Schorfheide (2007)

[ekaya]

thanks again