Well, reading the code is not much harder than reading the papers (not that
that helps much, been there...)
I don't actually know the answer, but the notation comes from this bit in
ca.jo():
if (spec == "longrun") {
ZK <- cbind(x[-c((N - K + 1):N), ], 1)
Lnotation <- K
}
else if (spec == "transitory") {
ZK <- cbind(x[-N, ], 1)[K:(N - 1), ]
Lnotation <- 1
}
colnames(ZK) <- c(paste(colnames(x), ".l", Lnotation,
sep = ""), "constant")
(actually there are several such bits).
K=2 by default, so you get .l2 for the "longrun" spec and ".l1" for
"transitory". So I would guess that studying the two specification formats from
the help page might give the solution to the riddle eventually.
(Another issue is that the column names are clearly rubbish, only the row names
make sense. The columns are eigenvectors sorted by eigenvalues which has no
relation to the input columns. Presumably, they are just an artifact of the
matrix operations.)
-pd
> On 24 Aug 2018, at 11:31 , Ashim Kapoor <ashimkap...@gmail.com> wrote:
>
> Dear Bert,
>
> I have read some of the references. I do understand what the 2 matrices(
> the cointegrating relationships and the alpha / loading matrix which gives
> the speed of the mean reversion) are. What I do not understand is the
> format of the output of the package. My main query is that why do we have
> .l2 in the cointegrating relationships. They are contemporaneous
> relationships , they should not have .l2 in the end. That's my query.
>
> Many thanks,
> Ashim
>
> On Thu, Aug 23, 2018 at 7:52 PM Bert Gunter <bgunter.4...@gmail.com> wrote:
>
>> This is about statistics , not R programming, and so is off topic here.
>> Your first port of call for this sort of thing should be the package docs,
>> **including any references** . There are references given. Have you studied
>> them??
>>
>> Cheers,
>>
>> Bert Gunter
>>
>> "The trouble with having an open mind is that people keep coming along and
>> sticking things into it."
>> -- Opus (aka Berkeley Breathed in his "Bloom County" comic strip )
>>
>>
>> On Thu, Aug 23, 2018 at 2:12 AM Ashim Kapoor <ashimkap...@gmail.com>
>> wrote:
>>
>>> Dear All,
>>>
>>> I am not sure about the summary of the function ca.jo. I have posted my
>>> query here :-
>>>
>>>
>>> https://stats.stackexchange.com/questions/363188/interpreting-the-names-used-in-the-output-of-johansen-test-in-package-urca-in-r
>>>
>>> I did not receive any reply so I am posting my query here.
>>>
>>> Many thanks and best regards,
>>> Ashim
>>>
>>> [[alternative HTML version deleted]]
>>>
>>> ______________________________________________
>>> R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see
>>> https://stat.ethz.ch/mailman/listinfo/r-help
>>> PLEASE do read the posting guide
>>> http://www.R-project.org/posting-guide.html
>>> and provide commented, minimal, self-contained, reproducible code.
>>>
>>
>
> [[alternative HTML version deleted]]
>
> ______________________________________________
> R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
--
Peter Dalgaard, Professor,
Center for Statistics, Copenhagen Business School
Solbjerg Plads 3, 2000 Frederiksberg, Denmark
Phone: (+45)38153501
Office: A 4.23
Email: pd....@cbs.dk Priv: pda...@gmail.com
______________________________________________
R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
