Error variance specification in State-space models

For technical questions regarding estimation of single equations, systems, VARs, Factor analysis and State Space Models in EViews. General econometric questions and advice should go in the Econometric Discussions forum.

Moderators: EViews Gareth, EViews Moderator

ErikG
Posts: 78
Joined: Wed Jan 23, 2013 1:18 am

Error variance specification in State-space models

Postby ErikG » Mon Dec 28, 2020 8:08 am

Hi,

Why are the variances for the errors in equations in State Space models specified as:

@state ... + [var=exp(c(1))]

and not just

@state ... + [var=c(1)]

is exp() here the exponential function? Is it to ensure non-negative variances?

Should one take the natural logarithm of the the estimated c(1) parameter in the first case in order to obtain the extimated variance of the error term?

Thanks,
Erik

startz
Non-normality and collinearity are NOT problems!
Posts: 3775
Joined: Wed Sep 17, 2008 2:25 pm

Re: Error variance specification in State-space models

Postby startz » Mon Dec 28, 2020 8:45 am

ErikG wrote:Hi,

Why are the variances for the errors in equations in State Space models specified as:

@state ... + [var=exp(c(1))]

and not just

@state ... + [var=c(1)]

is exp() here the exponential function? Is it to ensure non-negative variances?

Yes.
Should one take the natural logarithm of the the estimated c(1) parameter in the first case in order to obtain the extimated variance of the error term?

Nope, the estimated variance is exp(c(1)).


Return to “Estimation”

Who is online

Users browsing this forum: Google [Bot] and 18 guests