Rolling forecast with fixed estimation window
by pirre andersson
Hi, im new to gretl and this mail list but i thought i'd give it a chance. Im currently doing a comparison between recursive and rolling forecats with h-step ahead (1,3,6) The script im trying to use i found in an earlier mail from the archive and i tried to modify it abit to suit my needs.currently it looks like open kandidat.gdt
scalar DataLen = $nobs
scalar WindowLen = 191
scalar FirstObs = 1
scalar LastObs = WindowLen
matrix FCMat = zeros(DataLen, DataLen - LastObs)
scalar j = 1
loop t = LastObs + 1..DataLen - 1
smpl FirstObs LastObs
ols OMX30 const OMX30(-1) OMX30(-2) OMX30(-3) OMX30(-4) OMX30(-5) OMX30(-6) OMX30(-7) OMX30(-8) OMX30(-9) OMX30(-10) OMX30(-11) OMX30(-12) --robust --quiet
smpl 1986:2 2002:12
fcast FSeries (#is it possible to add --rolling and --out-of-sample here)
FCMat[,j] = FSeries
FirstObs = FirstObs + 1
LastObs = LastObs + 1
j++
endloop > ># drop missing values and print >FCMat = FCMat[4:,]print FCMat What i want to do is for example to try and use 1986:2-2002:12 as the initial estimation period and roll the window forward while dropping the first obsvervation for each step i move "forward". Then do out of sample forecasts for 2003:1 to 2012:12 while keeping the estimation window rolling at 191observations and doing either 1,3 or 6 steap ahead forecast. Is this possible and in that case Best regards
Pierre
12 years, 1 month
Multiple IRFs in the same graph with C.I.
by Gabriela Nodari
Dear gretl users,
I am dealing with a Svar model and I have plotted some impulse responses,
obtained while doing a sensitivity analysis, all in the same graph.
However, I would like to plot also the corresponding confidence intervals.
I have some idea about how to plot them with lines, but I would like to
keep the baseline confidence interval as a shaded area, as it is by
default.
Is there a way to do this?
Thank you.
Gabriela
12 years, 1 month
$irf(): excess collinearity
by artur tarassow
Dear gretl programmers,
I've got an obvious issue with collinearity in my model. When I try to
compute IRFs with bootstrapped confidence intervals, gretl stops and gives
the following message:
"Excessive collinearity in resampled datasets".
This is a good hint, no question. But what is the specific criteria for
gretl to stop here? Furthermore, despite of setting "set halt_on_error
off", gretl stops.
Btw, is there a way to obtain the analytical standard errors for the IRFs?
Thank you!
Artur
12 years, 1 month
MAP & MAPE in fcstats()
by yinung＠Gmail
Dear all,
I am wondering if MAPE and MPE shown in fcstats are incorrectly placed.
Please try the following script to compare MAPE and MPE calculated manually
and by fcstats.
Thanks
Yi-Nung Yang
<hansl>
nulldata 8
matrix x = {1,2,3,4,5,6,7,8}
matrix xhat={4,4,4,4,4,4,4,4}
series y=x
series yhat=xhat
MAPE=100*sum(abs(y-yhat)/y)/8
MPE=100*sum((y-yhat)/y)/8
mm=fcstats(y,yhat)
print mm
</hansl>
12 years, 1 month
Uncentered R-squared in case of non-intercept model
by Anton Pyzhev
Dear colleagues,
First of all, many thanks for gretl. I am using it with my students and research activities since last year, and I do really enjoy it.
One small issue raised concerning the goodness-of-fit test for OLS models which don't have the intercept term. As far as I understand, gretl computes the common R-squared for intercept models, and uncentered R-squared. It is known that in such models the uncentered is very likely to be close to 1, hence it becomes uninformative.
Would it be possible to include the option for choosing, what kind of R-squared will be computed for non-intercept models?
Best regards,
Anton Pyzhev,
Assistant Professor
at the Institute of Economics, Management and Environmental Studies,
Siberian Federal University,
Krasnoyarsk, Russia
12 years, 1 month
Gretl crashes
by Manny B
Hi,Lately, I notice a problem with Gretl. It constantly crashes when moving (left-right) for example OLS specify model panel. I use Windows..if this information helps.Thnx.Manny
12 years, 1 month
yule walker
by aymen kaabi
hello
CAN I USE YULE WAKLER WITH GRETL VAR ?
IF YES HOW CAN I DO
THANKS FOR YOUR HELP
12 years, 2 months
Gretl works :)
by Manny B
What happened Allin?! Today s snapshot works perfectly :))) Woohooo :)Manny
12 years, 2 months
Re: [Gretl-users] Gretl crashes
by Dr RJF Hudson
Dear Sven,
Yes thanks Sven, I am aware that support for XP ends soon, thanks.
Informatively I've come from using an Apple IIe and Apple Soft and DOS 6.22
through IBM and XT's and W95 and .......to XP Pro...through Verbatim 5 1/4
inch
and on to to 3 1/4 inch floppies and you name it. 10 meg external HDD's
and 6205 chips were wondrous..!
I'm such a fan of the reliability and stable horsepower of XP that
I've purchased a new sealed copy of XP SP3, as backup.
I've just never had a crash Sven, so I guess just as you say, I've been
lucky.
As we speak I'm generating (simulating) 500 data files each file having 225
obs
from which I'll be generating eq A6.2 through to A6.14 Pyndick and
Rubinfeld.
My machine doesn't even blink at the job..
I wish you well, Sven. Work hard and thankyou very much for your post...
----- Original Message -----
From: "Sven Schreiber" <svetosch(a)gmx.net>
To: <gretl-users(a)lists.wfu.edu>
Sent: Monday, May 06, 2013 6:35 PM
Subject: Re: [Gretl-users] Gretl crashes
> Am 04.05.2013 11:13, schrieb Dr RJF Hudson:
>> Greetings all,
>> Please might I give two bob's worth of support for XP Pro.
>
> I agree XP should still be supported by gretl. However, you are aware
> that official MS support for its own product ends in less than a year,
> right?
>
>> I have found it to be very very stable. It runs just about everything I
>> throw at
>> it, from compiled F90 through R and Gretl and 2 screens, to compiled
>> GWBasic....programs with big powerful recursion, too, generating
>> very big matrices..
>> XP has never ever crashed with me of its own volition.
>
> Well, I guess you have been lucky...
>
> cheers,
> sven
>
>
>
> _______________________________________________
> Gretl-users mailing list
> Gretl-users(a)lists.wfu.edu
> http://lists.wfu.edu/mailman/listinfo/gretl-users
12 years, 2 months
Gretl crashes
by JOSE FRANCISCO PERLES RIBES
Dear all:
I have tried the cvs in a portable computer Intel Pentium Processor 1,73
GHz and Windows XP Profesional SP3. Result: Gretl runs OK.
Best regards
José Perles
University of Alicante
12 years, 2 months
