DSGE growth model and detrending

#1

Dear users,

Do I have to detrend all my variables (variables expressed in labor efficiency units) to simulate a growth model? Or can I just express them in per capita?

Data and the model should be detrended in the same way, right?

Thank you in advance.

0 Likes

#2

You need eleborate. What exactly are you trying to do? Estimation? Or just a simulation? If the latter, do you do a stochastic simulation or perfect foresight?

0 Likes

#3

Thank you for your answer.

I am trying to do stochastic simulations of my model (and some forecasting). I also have some parameters to estimate from some observable variables.

Best regards

0 Likes

#4

For stochastic simulations, your model needs to have a well-defined steady state. So you cannot have growth trends in the model, i.e. you need to write down a model with well-defined BGP and add the trend back later, e.g. by relying on first differences. The same applies for estimation, where you need to use a stationary transformation of the data.

0 Likes

#5

That’s what I thought: detrending the model and the data.

So if I want nice graphs exhibiting growth trends, I have to do a deterministic simulation of my model ?

Thank you in advance.

Have a nice weekend

0 Likes

#6

No, you need to work with the detrended model and later add the trend back. That is straightforward.

0 Likes

#7

OK thank you very much.

Do you know a textbook for derivating the BGP with nested-production functions in an economy? I am having a hard time doing it and cannot find anything on the net.

Best

0 Likes

#8

Unfortunately, I am not aware of such a reference.

0 Likes

#9

Thank you.

One last question: if I simulate my model from 2017 to 2035, will my steady state represent year 2035 (or 2017?), since it is the long-run state of my economy? I calibrate the values of my parameters on year 2017 representing a near to zero output gap.

0 Likes

#10

“long-run” in this context (a linear model) means not a particular year, but rather the long-run average over time or where the economy is if there are no shocks happening.

0 Likes

#11

Good evening,

I am finally working on a perfect-foresight model. In that case the model does not have to be stationary, right? So I can put my model without detrending. In that case, do I still need to specify growth trends (I have two of them)?

What about bayesian estimation? Do I have to transform the data?

Thanks

0 Likes

#12
  1. No, perfect foresight models need not be stationary. See e.g. https://github.com/JohannesPfeifer/DSGE_mod/blob/master/Solow_model/Solow_nonstationary.mod
  2. For Bayesian estimation, you cannot use perfect foresight. Both model and data need to be stationary.
0 Likes

#13

Oops… So how do I estimate my elasticities of substitution if I cannot use bayesian estimation?

Thanks

0 Likes

#14

Without context that is hard to tell. Why can’t you detrend your model?

0 Likes

#15

I have two different technical progress, so my model won’t be stationary in Uzawa’s sense. I’m gonna try to shut down one technical progress, so I only have labor-augmenting technical progress and can stationnarise the model. I think it’s best.

Can I use perfect foresight with a stationary model and perform a Bayesian estimation? Or does it have to be in a stochastic environnement necessarily?

Thank you

0 Likes

#16

Bayesian estimation requires a stochastic model.

0 Likes

#17

Damn…

Well. I’m going to try to simplify my model as much as I can then.

Thank you.

Have a nice WE

0 Likes

#18

Is it the same for maximum likelihood estimation? Does it require the model in stationary form?

Thank you

0 Likes

#19

Yes. As soon as you want to construct a likelihood function, you need a stochastic model.

0 Likes