Question about Identification?


#1

Dear Professor Pfeifer,

I have ecountered a problem when doing the Bayestion estimation. Would you please give some advice? Thank you very much.

(1) When I put ‘‘identification;’’ before estimation, the result displays as

==== Identification analysis ====

Testing current parameter values
The number of moments with non-zero derivative is smaller than the number of parameters
Try increasing ar = 2
The number of moments with non-zero derivative is smaller than the number of parameters
Try increasing ar = 3
The number of moments with non-zero derivative is smaller than the number of parameters
Try increasing ar = 4
The number of moments with non-zero derivative is smaller than the number of parameters
Try increasing ar = 5
The number of moments with non-zero derivative is smaller than the number of parameters
Try increasing ar = 6
The number of moments with non-zero derivative is smaller than the number of parameters
Try increasing ar = 7
The number of moments with non-zero derivative is smaller than the number of parameters
Try increasing ar = 8
The number of moments with non-zero derivative is smaller than the number of parameters
Try increasing ar = 9
The number of moments with non-zero derivative is smaller than the number of parameters
Try increasing ar = 10
The number of moments with non-zero derivative is smaller than the number of parameters
up to 10 lags: check your model
Either further increase ar or reduce the list of estimated parameters

What does this mean and what should I do to improve it ?

And today when I run the code in dynare 4.5.6 (the results above are in dynare 4.4.3), it displays:
There are NaN’s in the theoretical moments: make sure that for
non-stationary models stationary transformations of
non-stationary observables are used for checking identification.
[TIP: use first differences].

What does this mean?

(2) If I want to choose labor ( L in Y=AK^alphaL^(1-alpha) ) as observed variable, which data should I use? Is the number of employments appropriate?

Any reply will be appreciated. Thank you!


#2
  1. I would stick with 4.5.6. That message is more informative. The issue is that some of your observables are non-stationary, i.e. have a unit root. In that case, you need to use a stationary transformation as an observable if you want to check identification.
  2. The number of employees is only a rough proxy for hours worked as you are neglecting the intensive margin. But if nothing else is available, you should use it together with measurement error.