ARDL Bounds Testing programming

[londonphd]

Hello,

I am trying to learn how to program a model of ARDL approach to cointegration in Eviews, I want to do a rolling cointegrating analysis. There is a limited information on how to program an ARDL model and estimate long run and ECM equations. I manged to execute the following only:

Code: Select all

ardl(deplags=4, reglags=4) y x1 x2 x3 x4

I want to save the long run equation coefficients in a matrix, and conduct a bounds test then save the F-stat in a new series. Can anyone help with coding please?
Thank you

[EViews Gareth]

You'll have to name the equation you're estimating, then use equationname.bounds test to perform the bounds test.

Equationname.cointrep performs the countegration/ long run viee

[londonphd]

Thanks Gareth!

[londonphd]

Gareth,

Another quick question. I am reading the online manual which has more instructions than the pdf commands manual. But, I've been unable to save the f-stat from the bounds test and long run coefficient estimates. How can I save these in, let's say, a matrix?


Thanks

[EViews Gareth]

The PDFs and online help are identical.

viewtopic.php?f=5&t=18

[londonphd]

Gareth,

They are not the same, maybe I have one of many pdf manuals. I found the online help more detailed and your thread on programming helped me to get around the problems so far.

Also, can I call Eviews from within R?

Thanks a lot!

[EViews Gareth]

No

[fmramos]

Hi Gareth and EV users!

Expanding this topic, I have an issue. As a matter of fact, a doubt:

Is it possible to extract/to create/get a vector with long term / cointrep terms?

Extract results Bi's from: Cointeq = LOG(Y) - (B1*LOG(X1) + B2*LOG(X2) )?

Tks in advance.

Best,

Fabio

[EViews Gareth]

equation.@ardlcoint

[fmramos]

Tks. It worked to get all parameters from d(y)~d(x) relationship. But I'm trying to get directly 'long run coefficients', y~x relationship (inside cointreq equation) and long-run adjustment term (cointeq(-1)) parameter.

As like:

Cointeq = Y - (b1*X1 + b2*X2 -a1) from long-run form and period-to-period adjustment term parameter, g1 from 'g1*cointeq'.

Best,

Fabio

[EViews Gareth]

Ah, no.

[fmramos]

: (

No problem! Tks anyway for prompt answering. As always.

Best,

Fabio

[srisasoft]

I am trying to learn how to program a model of ARDL approach to cointegration in Eviews, I want to do a rolling cointegrating analysis. There is a limited information on how to program an ARDL model and estimate long run and ECM equations. I manged to execute the following only:

First, it's important to note that although there was previously an EViews "add-in" for ARDL models (see here and here), this was quite limited in its capabilities. What's now available is a full-blown ARDL estimation option, together with bounds testing and an analysis of the long-run relationship between the variables being modelled.