10:45 a.m.
Hi,
the current behaviour of fracdiff when invoked with a series containing NAs is to issue an
E_DATA error. I propose that in this case fracdiff returns a series of NAs. I think
returning NAs would be more in line with the rest of gretl, but I'm not sure.
The attached patch changes the fracdiff function in the proposed way.
Christoph
11:23 a.m.
On Sat, 20 Jun 2009, Klein, Christoph wrote:
the current behaviour of fracdiff when invoked with a series
containing NAs is to issue an E_DATA error. I propose that in
this case fracdiff returns a series of NAs. I think returning
NAs would be more in line with the rest of gretl, but I'm not
sure. The attached patch changes the fracdiff function in the
proposed way.
Thanks. Jack's away from home right now; since fracdiff is his
baby I think I'll wait and let him make the judgement on that.
This is a tricky point. I'm not sure we're entirely consistent in
what we do at present (fracdiff apart).
Allin
11:01 a.m.
New subject: NA patch for fracdiff functions
On Sat, 20 Jun 2009, Allin Cottrell wrote:
On Sat, 20 Jun 2009, Klein, Christoph wrote:
> the current behaviour of fracdiff when invoked with a series
> containing NAs is to issue an E_DATA error. I propose that in
> this case fracdiff returns a series of NAs. I think returning
> NAs would be more in line with the rest of gretl, but I'm not
> sure. The attached patch changes the fracdiff function in the
> proposed way.
Thanks. Jack's away from home right now; since fracdiff is his
baby I think I'll wait and let him make the judgement on that.
I think Christoph is right in that one NA in the series shouldn't stop
execution. However, since fractional differencing is a one-sided filter, I
think we may happily carry on until the first NA is found; this may be
especially useful if you've got a series where the missing observation is
near the end. I just committed a smoll modification to lib/src/genfuncs.c
to that effect. Please review!
Riccardo (Jack) Lucchetti
Dipartimento di Economia
Università Politecnica delle Marche
r.lucchetti(a)univpm.it
http://www.econ.univpm.it/lucchetti
1:39 p.m.
New subject: AW: NA patch for fracdiff functions
I checked out from cvs and it works fine now! Thanks!!
Christoph
________________________________________
Von: gretl-devel-bounces(a)lists.wfu.edu [gretl-devel-bounces(a)lists.wfu.edu] im Auftrag von
Riccardo (Jack) Lucchetti [r.lucchetti(a)univpm.it]
Gesendet: Mittwoch, 1. Juli 2009 17:01
An: Gretl development
Betreff: Re: [Gretl-devel] NA patch for fracdiff functions
On Sat, 20 Jun 2009, Allin Cottrell wrote:
On Sat, 20 Jun 2009, Klein, Christoph wrote:
> the current behaviour of fracdiff when invoked with a series
> containing NAs is to issue an E_DATA error. I propose that in
> this case fracdiff returns a series of NAs. I think returning
> NAs would be more in line with the rest of gretl, but I'm not
> sure. The attached patch changes the fracdiff function in the
> proposed way.
Thanks. Jack's away from home right now; since fracdiff is his
baby I think I'll wait and let him make the judgement on that.
I think Christoph is right in that one NA in the series shouldn't stop
execution. However, since fractional differencing is a one-sided filter, I
think we may happily carry on until the first NA is found; this may be
especially useful if you've got a series where the missing observation is
near the end. I just committed a smoll modification to lib/src/genfuncs.c
to that effect. Please review!
Riccardo (Jack) Lucchetti
Dipartimento di Economia
Università Politecnica delle Marche
r.lucchetti(a)univpm.it
http://www.econ.univpm.it/lucchetti
5618
days inactive
5629
days old
3 comments
3 participants
participants (3)
-
Allin Cottrell -
Klein, Christoph -
Riccardo (Jack) Lucchetti