I am new to gretl.

I would be very grateful to someone who could explain me how to predict a VAR model...

Datum: Wed, 18 Nov 2009 18:59:42 +0200 Von: Remigijus Lapinskas <remigijus.lapinskas(a)mif.vu.lt&gt; An: Gretl list <gretl-users(a)lists.wfu.edu&gt; Betreff: [Gretl-users] change frequency

Dear all, I have daily (5 days) observations. Is it possible to create a new weakly series of every Monday data? Many thanks, Rem _______________________________________________ Gretl-users mailing list Gretl-users(a)lists.wfu.edu http://lists.wfu.edu/mailman/listinfo/gretl-users

It might also be worth noting that you can do this sort of thing (or generalizations of it) easily with basic script commands. <script> nulldata 60 # fake daily data setobs 5 1 --time-series series x = normal() genr time genr mondum = (time % 5 == 1) print time mondum x -o smpl mondum --dummy setobs 52 1 --time-series print time mondum x -o store weekly.gdt </script>

Datum: Wed, 18 Nov 2009 22:59:37 -0500 (EST) Von: Allin Cottrell <cottrell(a)wfu.edu&gt; An: Gretl list <gretl-users(a)lists.wfu.edu&gt; Betreff: Re: [Gretl-users] change frequency

On Thu, 19 Nov 2009, yinung at Gmail wrote: > How can I generate a Monday dummy if the daily data is on a irregular > week base, for example, djclose.gdt? You can't expect this to be super-simple, can you? But here goes... <script> function scalar day_of_week (scalar yr, scalar mo, scalar day) # returns 0 = monday, 1 = Tuesday, ... # See Uspensky and Heaslet, Elementary Number Theory (1939) scalar c, d, i1, i2, i3 if (mo < 3) yr-- mo += 10 else mo -= 2; endif c = yr / 100 d = yr % 100 i1 = int(floor(2.6 * mo - 0.2) % 7) i2 = int(floor(d / 4.0) % 7) i3 = int(floor(c / 4.0) % 7) return ((day%7)+i1+(d%7)+i2+i3-((2*c)%7))%7 end function open djclose.gdt scalar T = $nobs scalar yr, mo, day string datestr series monday = 0 set messages off loop i=1..T -q smpl i i outfile date.txt --write print djclose -o outfile --close string date = $(tail -n+4 date.txt | head -1 | awk '{print $1}') sscanf date, "%d/%d/%d", &yr, &mo, &day printf "yr = %d, mo = %d, day = %d\n", yr, mo, day scalar dow = day_of_week(yr, mo, day) if dow == 0 monday[i] = 1 endif endloop smpl --full print djclose monday -o </script> If you're on MS Windows and don't have the basic text-processing tools tail, head and awk, go to gnuwin32.sourceforge.net for salvation. Allin Cottrell _______________________________________________ Gretl-users mailing list Gretl-users(a)lists.wfu.edu http://lists.wfu.edu/mailman/listinfo/gretl-users

Datum: Thu, 19 Nov 2009 09:21:16 -0500 (EST) Von: Allin Cottrell <cottrell(a)wfu.edu&gt; An: Gretl list <gretl-users(a)lists.wfu.edu&gt; Betreff: Re: [Gretl-users] change frequency

On Thu, 19 Nov 2009, Sven Schreiber wrote: > Isn't djclose.gdt just a 5-day weekly dataset? I mean, menu Add > -> periodic dummies would seem to do the trick -- I suspect > there was a misunderstanding about the term "irregular", no? No, djclose has holidays missing. If you construct the Monday dummy as I indicated you'll see that it's not strictly periodic.

In addition, since this is a bit tricky, I've added a built-in function "weekday" (in CVS and snapshot) which does the calculation given year, month and day.

On Fri, 20 Nov 2009, Riccardo (Jack) Lucchetti wrote: > On Thu, 19 Nov 2009, Allin Cottrell wrote: > >> In addition, since this is a bit tricky, I've added a built-in >> function "weekday" (in CVS and snapshot) which does the >> calculation given year, month and day. Sorry guys, I hit "Send" by mistake. What I wanted to say is that this is a very useful function. I wonder if we ought to have more date functions, like for example functions for dealing with the excel convention for representing dates (which, I believe, sets 0 to 1/1/1901, but I may be wrong).

It would be great if we can have a weeksay fundtion working for (irregular) daily data as follows series wkDummy1 = weekday()

Remigijus Lapinskas schrieb: > Dear all, > > I have two columns (irregular frequences): > > x y > > 0 2.1 > 0 3.4 > > 1 5.0 > 1 1.1 > 1 2.2 > > 2 3.2 > > 3 1.2 > 3 8.8 > .... > > I want to average y's in the first group (i=0), then second(i=1) etc How > can I do this? > matrix X = {x, y} # if x, y are series matrix myaverages = meanc(X)

Sven, Thank you for your swift replay, but this not exactly what I want. In my case I want to get a vector (2.1+3.4)/2, (5.0+1.1+2.2)/3, 3.2/1, (1.2+8.8)/2 etc

Hi Rem, That's a grouped mean that you need I guess. Try this: smpl x=0 --restrict summary y smpl x=1 --restrict --replace summary y smpl x=2 --restrict --replace summary y smpl x=3 --restrict --replace summary y artur Remigijus Lapinskas a écrit : > Sven, > > Thank you for your swift replay, but this not exactly what I want. In my > case I want to get a vector (2.1+3.4)/2, (5.0+1.1+2.2)/3, 3.2/1, > (1.2+8.8)/2 etc > > Cheers, > Rem > > > Sven Schreiber wrote: > >> Remigijus Lapinskas schrieb: >> >>> Dear all, >>> >>> I have two columns (irregular frequences): >>> >>> x y >>> >>> 0 2.1 >>> 0 3.4 >>> >>> 1 5.0 >>> 1 1.1 >>> 1 2.2 >>> >>> 2 3.2 >>> >>> 3 1.2 >>> 3 8.8 >>> >>> >>> I want to average y's in the first group (x=0), then second(x=1) etc How >>> can I do this? >>> >>> >> matrix X = {x, y} # if x, y are series >> matrix myaverages = meanc(X) >> >> cheers, >> sven >> _______________________________________________ >> Gretl-users mailing list >> Gretl-users(a)lists.wfu.edu >> http://lists.wfu.edu/mailman/listinfo/gretl-users >> >> >> > _______________________________________________ > Gretl-users mailing list > Gretl-users(a)lists.wfu.edu > http://lists.wfu.edu/mailman/listinfo/gretl-users > > > _______________________________________________ Gretl-users mailing list Gretl-users(a)lists.wfu.edu http://lists.wfu.edu/mailman/listinfo/gretl-users