Re: [Gretl-users] problems with daily data
Am 21.05.2014 11:49, schrieb Riccardo (Jack) Lucchetti:
On Wed, 21 May 2014, Sven Schreiber wrote:
> Next I would like to convert the workfile from 7-days-daily to
> 5-days-daily, i.e. remove the empty (=missings) weekends obs, but this
> doesn't seem to be easy to do. (Note that removing all missing obs is
> not equivalent, because that also eliminates other holidays, which is
> another topic. Also I know I could use 'join' together with the source
> csv file to do that, but I think it should be easier than that, no?)
This should do what you want: not the most elegant approach, but IMO
quite clear and general. The only thing I don't like very much is the
necessity to go through a temporary file, which I find quite ugly.
Allin, we talked a bit about the possibility of having a "forever"
option to the smpl command, which would effectively eliminate certain
rows from a dataset (so that operations like "keep" and "drop" in
Stata)
become quite easy. What do you think?
I'm not Allin, but also related to the issue of the other holidays in
the data, this would be helpful I think. I would guess that people
working with daily data (which I rarely do) treat the time series as
being equally spaced over time (=as a standard time series) even though
it is not, strictly speaking, due to the weekends and holidays. I mean
that's what gretl also does with its 5-day-daily and 6-day-daily formats.
But currently in gretl I'm always losing the time-series property of my
workfile when I remove the missing holidays, and so I could not use the
built-in time series tools. My workaround has been to save this as a
special time series workfile with periodicity 1, so no date information
anymore. But this feels like a cludge (for example no daily dummies,
except if I created them beforehand).
/*
trash weekends
*/
# first, construct a "weekend" dummy series
scalar y1 = $obsmajor[1]
scalar m1 = $obsminor[1]
scalar d1 = $obsmicro[1]
scalar wd1 = weekday(y1, m1, d1)
series wd = time + wd1 - 1
series we = (wd%7)==6 || (wd%7)==0
Thanks Jack; I tried this more compactly with:
<hansl>
genr dummy # BTW, need a new syntax for this IMHO
series we = dummy_6 || dummy_7
</hansl>
but apparently the daily dummies are allocated "randomly", i.e. dummy_1
starts with whatever is the first obs in the workfile (or sample?), not
Monday or Sunday. Could this be changed perhaps?
thanks,
sven