EViews.com

Perfect - thanks Gareth.

Hi, I would like to create a user input box like what pops up when you select series -> Unit root test.. I had a quick look at the documentation, and I couldn't find an example of how to create a input box that combines things like list and radio buttons. Could someone please point me in correct dir...

I don't think that is the intention, no. The original specification was: Y = F*X(t) + E(t) (X(t)-U) = G*(X(t-1)-U) + W(t) In R, the state space model doesn't allow for an intercept in the state equation (which the above has) so to get around that I have moved the intercept into the state vector: X(t...

Thanks Glenn, indeed an error. In regards to the state equations, I have specified this model a little differently to how I have seen others on this forum specify this model. This is because I originally estimated the model using R's DLM package, which doesn't allow an intercept in the state equatio...

Hi,

I've attached a workfile if it helps - any guidance would be appreciated.

Adam

Hi, I am trying to convert a model I had put together using R's DLM (Dynamic Linear Model) into Eviews Sspace framework. The model is a Nelson Siegal style latent factor model, using inflation expectations rather than bond yields. I can't seem to get estimates anywhere near what I get with R. I've p...

Hello,

A really nice addition would be a key board shortcut (like CTRL+ENTER or something) for the right click command "Run selected".

Thanks

Adam

Hi, Is there any difference between a) creating a coefficient vector using the command "coef" and then setting the values equal to desired starting values and b) using the command param? Specifically, is coef(3) a a(1) = 0.1 a(2) = 0.5 a(3) = -1 treated differently to coef(3) a param a(1) ...

Couldn't you re-specify your dependent variable to Y-X1?

Hi,

A really neat addition would be if program editor had parenthesis highlighting, indicating which parenthesis closes with which.

Thanks

Adam

Hi Matt,

Just closing the loop on this. Turns out your initial hunch was correct, there were some add-factors that were dropped at some stage. Using the addassign before the addinit command does actually work - must of bungled it when I first was experimenting.

Cheers,

Adam

Thanks for the suggestion Matt - I haven't checked if all variables are different, just some key ones. I'll try and find the equation where things fall over.

And, yep totally understand it is hard to offer advice without seeing the workfile. But thanks for helping where you can :)

Thanks for the reply Matt. I thought somethings similar, I used the command model.addassign(n) @all model.addinit(v = n) @all Which I was hoping would create zero filled add-factors for all equations in the model, then addinit setting so there are no residuals at actuals. Just as an aside, does my a...

Hi, I've come across a bit of a peculiar problem, which I'll do my best to describe. In coming up with a baseline for my model, a few variables are excluded and overridden, some add-factors are adjusted etc. I then want to run scenarios off this baseline, but in the scenarios I want the full functio...

Hi, In Chapter 2 Durbin and Koopman's Time Series Analysis by State Space Methods , the authors demonstrate that one of the variance for the local level model - the state variance - can be concentrated out of the likelihood. What the authors call the "Concentrated diffuse loglikelihood" is...