Missing observations

[gabrielbruneau]

Hello,

I need information about the treatment by Dynare of the missing observations for estimation.

Dynare removes them, along with all observations of the other variables for the same time period?
Dynare removes them, and keep all observations of the other variables for the same time period?

I ask this question for the following reason: I want to estimate a DSGE model with ten observables: Eight of them cover 1976Q1 to 2012Q1, and two of them cover 1983Q1 to 2012Q1. If I include the observations for the eight variables that go back to 1976 and include NaN for the two variables that go back to 1983 from 1976 to 1982, will Dynare estimate the model from 1976 to 2012, will Dynare estimate the model from 1983 to 2012?

Thanks
Gabriel

[jpfeifer]

Dynare will use the longer sample and treat the missing values as unobserved states during the Kalman filter routine, i.e. their best forecast is inferred from the model and the remaining data.

[gabrielbruneau]

Thank you very much for your answer, it is really helpful to know that information.

Gabriel