Information about Detrending Variables

Estimation ML/Bayesian estimation
1

Dear Professor Johannes Pfeifer

I have read your excellent paper titled “A Guide to Specifying Observation Equations for the Estimation of DSGE Models”, University of Cologne, July 8, 2018. I want to ask you the following question, please:

In section 1.1 A Baseline RBC Model (page, 7), equation (2), you set labor hours, h_t=1. Imagine, please, that we have a model in which h_t\neq 1. In this case do we still have to divide our observable variables by the civilian labor force? For instance, set Y_{\text{pc},t}^{\text{data}}= Y_t^{\text{data}}/\text{N}_t^{\text{data}} ?.

Thank you very much for your answer.

Kind Regards

Sinclair

2

The quantities in the theoretical model are defined in per capita terms. Hence, the division of the data series by population or labour force. Yes, this is true even when hours worked are not set to one.
Reuben

3

Dear Professor Reuben

Thank you very much for your answer.

Kind Regards

Sinclair