2:13 a.m.
Hello there,
I still have a simple coding question...
So essentially, I have a table with 20 columns and 6000 rows. Each row
represents a volatilty value for a specific time and I have this data for
~20 stocks in my specific universe. My question is, how do I create a loop
in GRETL where I run the Engle-Granger test on all possible permutations
for each Column (each stock name) and output a column with each pair (for
example Column A and Column B) and the level of co-integration (1- pvalue).
Ultimately, I would like to be able to see which combination (of stock
volatilities) are the most cointegrated.
Thanks again and I really hope to hear from you soon!
-A
2:39 a.m.
On Thu, 20 Sep 2012, Aditya Nemani wrote:
I still have a simple coding question...
So essentially, I have a table with 20 columns and 6000 rows. Each row
represents a volatilty value for a specific time and I have this data for
~20 stocks in my specific universe. My question is, how do I create a loop
in GRETL where I run the Engle-Granger test on all possible permutations
for each Column (each stock name) and output a column with each pair (for
example Column A and Column B) and the level of co-integration (1- pvalue).
Look at the loop examples in the manual and give it a try. Show us
what you came up with and you may get some suggestions for
improvements. But don't just say, "Do it for me".
Allin Cottrell
5:46 p.m.
Would something like this work?
matrix X
loop i=1..5 -q
loop m = i+1..5 -q
#Do stuff
endloop
endloop
Logan
6:39 p.m.
On Fri, 21 Sep 2012, Logan Kelly wrote:
Would something like this work?
matrix X
loop i=1..5 -q
loop m = i+1..5 -q
#Do stuff
endloop
endloop
As far as I'm correct in imagining what your intentions are (having a
nested loop in which the extremes of the inner loop depend on the counting
variable of the outer loop), yes, as demonstrated by the following
snippet:
<hansl>
loop i=1..5 -q
loop m = i+1..5 -q
printf "%d, %d\n", i, m
endloop
endloop
</hansl>
But, out of curiosity, wouldn't have taken you much less to try this by
yourself rather than asking the list? Note: it's good that you posted the
question to the list, so it automatically enters the realm of Google's
omniscience and others will find it in the future. I'm just asking.
--------------------------------------------------
Riccardo (Jack) Lucchetti
Dipartimento di Economia
Università Politecnica delle Marche
(formerly known as Università di Ancona)
r.lucchetti(a)univpm.it
http://www2.econ.univpm.it/servizi/hpp/lucchetti
--------------------------------------------------
6:43 p.m.
I was responding to an earlier request posted by anouther user? Should I have followed a
diffrent procedure?
T-Mobile, America's First Nationwide 4G Network
"Riccardo (Jack) Lucchetti" <r.lucchetti(a)univpm.it> wrote:
On Fri, 21 Sep 2012, Logan Kelly wrote:
Would something like this work?
matrix X
loop i=1..5 -q
loop m = i+1..5 -q
#Do stuff
endloop
endloop
As far as I'm correct in imagining what your intentions are (having a
nested loop in which the extremes of the inner loop depend on the counting
variable of the outer loop), yes, as demonstrated by the following
snippet:
<hansl>
loop i=1..5 -q
loop m = i+1..5 -q
printf "%d, %d\n", i, m
endloop
endloop
</hansl>
But, out of curiosity, wouldn't have taken you much less to try this by
yourself rather than asking the list? Note: it's good that you posted the
question to the list, so it automatically enters the realm of Google's
omniscience and others will find it in the future. I'm just asking.
--------------------------------------------------
Riccardo (Jack) Lucchetti
Dipartimento di Economia
Università Politecnica delle Marche
(formerly known as Università di Ancona)
r.lucchetti(a)univpm.it
http://www2.econ.univpm.it/servizi/hpp/lucchetti
--------------------------------------------------
6:48 p.m.
On Fri, 21 Sep 2012, Logan Kelly wrote:
I was responding to an earlier request posted by anouther user?
Should I
have followed a diffrent procedure?
Ah, ok, sorry! I hadn't seen the previous messages. Reading your message
on its own gave me the impression that you were asking a genuine question.
Sorry again.
--------------------------------------------------
Riccardo (Jack) Lucchetti
Dipartimento di Economia
Università Politecnica delle Marche
(formerly known as Università di Ancona)
r.lucchetti(a)univpm.it
http://www2.econ.univpm.it/servizi/hpp/lucchetti
--------------------------------------------------
4487
days inactive
4487
days old
5 comments
4 participants
participants (4)
-
Aditya Nemani -
Allin Cottrell -
Logan Kelly -
Riccardo (Jack) Lucchetti