Re: [Gretl-users] ADF by OLS manually
On Mon, 22 Apr 2013, yinung at Gmail wrote:
Sorry, it was a typo. In fact, it should be
ADF3 <- ols diff(Yt) const Yt(-1) diff(Yt(-1))
The result is even stranger. The last term was incorrectly parsed as
"diff(Yt)."
As things stand, nesting of transformations in the arguments
to a gretl _command_ (such as ols) is not supported. We just
support simple, single-level things such as "log" and "diff"
as a convenience. That's in contrast to function calls, where
transformations can be nested freely.
So the ols command line shown above should produce an error
message. That fact that it didn't -- the stuff inside diff()
just being truncated -- is a bug, now fixed in CVS.
You now get:
field 'diff(Yt(-1))' in command is invalid
Allin
--------------------------------------------------------------------------
gretl version 1.9.12cvs
Current session: 2013-04-22 21:49
? ADF3 <- ols diff(Yt) const Yt(-1) diff(Yt(-1))
ADF3: OLS, using observations 1960-1994 (T = 35)
Dependent variable: d_Yt
coefficient std. error t-ratio p-value
-------------------------------------------------------
const 0.00000 0.00000 NA NA
Yt_1 0.00000 0.00000 NA NA
d_Yt 1.00000 0.00000 NA NA
Mean dependent var 276.5429 S.D. dependent var 243.4384
Sum squared resid 0.000000 S.E. of regression 0.000000
R-squared 1.000000 Adjusted R-squared 1.000000
2013/4/22 Sven Schreiber <svetosch(a)gmx.net>
> Am 22.04.2013 10:03, schrieb yinung at Gmail:
>
>>
>> But this looks like Yt_1 has been generated internally since the script
>> like this does not work:
>>
>> ADF1 <- ols diff(Yt) const Yt(-1) diff(Y(-1))
>>
>
> well if your variable is called "Yt" (a bad choice IMHO BTW) then
"Y"
> isn't going to work, is it?
>
> In general it would help to include the error message.
>
> hth,
> sven
>
> _______________________________________________
> Gretl-users mailing list
> Gretl-users(a)lists.wfu.edu
> http://lists.wfu.edu/mailman/listinfo/gretl-users
>
--
Allin Cottrell
Department of Economics
Wake Forest University, NC