Fixed Effect crashed when I tried to include over 5000 entity dummies
by Fred Engst
> Hi Fred,
>
>> In either case, the buffer overflow should be looked at.
>
> Of course, gretl should not crash. Could you specify which gretl version
> on which OS you're using please? If you aren't using the latest stable
> version gretl 2019a, could you please try this one again.
>
> I can't say anything more on the Logit issue -- sorry.
>
> Best,
> Artur
>
Hi Artur,
here is the top part of my crash report:
------------------------------------------------------------------------------------
Process: gretl [10753]
Path: /Applications/Gretl.app/Contents/Resources/bin/gretl
Identifier: net.sourceforge.gretl
Version: 2019a-git
Code Type: X86-64 (Native)
Parent Process: ??? [1]
Responsible: gretl [10753]
User ID: 502
Date/Time: 2019-01-17 16:54:27.499 +0800
OS Version: Mac OS X 10.11.6 (15G22010)
Report Version: 11
Anonymous UUID: 76CACE37-6884-3AB3-F91C-F52618C62272
Sleep/Wake UUID: 0AD2F5AA-ECE2-4687-894D-0FE3111E22A0
Time Awake Since Boot: 160000 seconds
Time Since Wake: 30000 seconds
System Integrity Protection: enabled
Crashed Thread: 0 Dispatch queue: com.apple.main-thread
Exception Type: EXC_CRASH (SIGABRT)
Exception Codes: 0x0000000000000000, 0x0000000000000000
Exception Note: EXC_CORPSE_NOTIFY
Application Specific Information:
detected buffer overflow
-------------------------------------------------------------------------------------
I don’t suppose you need the rest of the report.
Fred
>
>> Just now I tried to run a fixed effect by including entity dummies on OLS through GUI, it crashed gretl once again, with the same crash report.
>
> Please clarify, do you mean you did a plain OLS estimation in panel data
> and adding a lot of unit dummies (literal fixed effects), and then it
> crashed? This would of course be a bug.
>
>> Next, I tried fixed effect by including du_* on the OLS command through gretlcli, it was able to complete in a few minutes, much longer than the builtin panel model with the ?fixed-effects option.
>
> This is not unexpected, because the standard way of almost all
> implementations is to demean the variables first (along the T-dimension,
> the within transformation). Instead adding the many unit dummies is
> theoretically possible but has numerical problems.
>
>> Is there a way to do the fixed effect like the builtin panel way rather than the explicit including dummies way on a tobit model?
>
> If you're prepared to do a biased estimation you might apply the within
> transformation to the variables manually and then estimate a standard
> Tobit model on the transformed variables. However, it's not clear to me
> how to preserve the zeros in the dependent variable then...
>
>> In either case, the buffer overflow should be looked at.
>
> Absolutely!
>
> thanks for the report,
> Sven
Hi Sven,
1. Yes, you are right. I did a plain OLS estimation in panel data and adding a lot of unit dummies and then it crashed.
The crash report is identical to the one above other than the crash date.
2. I didn’t realize that is how gretl does the fixed effect, for it reported LSDV R-squared, which I thought means it included the contribution of those dummies.
3. I really don’t know much or thought much about “fixed effect” tobit. I just thought by adding those dummies I don’t have to demean those dependent variables. Otherwise they will lose their non-nagetive nature, making tobit a misfit. Now I realized that to estimate over 5000 parameters is next to impossible anyway.
It seems that if I reduce the detentions of those “fixed effects” to a few characteristics, say less than 10, then the tobit model should work.
Thanks so much for your reply.
Fred
5 years, 2 months
No acess to dbnomics anymore
by klaus.hasenbach＠web.de
Hi,
I am quite new with gretl. In November I used gretl to retrieve data
from "dbnomics. world" sucessfuly. Great work done by the comunity!!
But now I get the message:
Fehlermeldung von dbn_get_json():
SSL peer certificate or SSH remote key was not OK
*** error in function dbnomics_providers, line 4
> err = dbn_get_json("providers", verbose, &json)
Any help? Thank you!
Klaus
5 years, 2 months
help needed
by valerie von arx
Hi guys,
I am using GRETL for my econometric analysis, which relies on gnu plot for any kind of graphic visualisation e.g. graphs/distributions etc.
I already worked in the past with gretl and it was working perfectly. Suddenly, today when i tried to graph something the software told me:
“ Gnuplot is broken or too old: must be >= version 5.0 “
I downloaded the new version of gnu plot for mac (which, by the way, should already be included in the gretl package, as the guide suggests) but I am not able to install it.
I tried to download other versions of both gretl and gnuplot but nothing changes.
I am really confused, I don’t know what else to try and I really need to get this analysis done.
Hope you could direct me on the right path.
Thank you in advance.
Val
5 years, 2 months
Re: [Gretl-users] Gretl-users Digest, Vol 144, Issue 29
by Stefano
I could not read the attachment so I cannot enter into the matter, but
as I understand it the question can be prephased as "why do the critical
values and the p-values computed by Gretl and R differ?". If I am
correct it is thus a perfectly legitimate question (indeed a good one).
bye
Stefano
Il 31/01/2019 18:00, gretl-users-request(a)lists.wfu.edu ha scritto:
> Send Gretl-users mailing list submissions to
> gretl-users(a)lists.wfu.edu
>
> To subscribe or unsubscribe via the World Wide Web, visit
> http://lists.wfu.edu/mailman/listinfo/gretl-users
> or, via email, send a message with subject or body 'help' to
> gretl-users-request(a)lists.wfu.edu
>
> You can reach the person managing the list at
> gretl-users-owner(a)lists.wfu.edu
>
> When replying, please edit your Subject line so it is more specific
> than "Re: Contents of Gretl-users digest..."
>
>
> Today's Topics:
>
> 1. Re: Discrepancy between R and Gretl in Johansen cointegration
> (Sven Schreiber)
> 2. Re: Discrepancy between R and Gretl in Johansen cointegration
> (Reynaldo Senra)
>
>
> ----------------------------------------------------------------------
>
> Message: 1
> Date: Thu, 31 Jan 2019 15:09:35 +0100
> From: Sven Schreiber <svetosch(a)gmx.net>
> To: gretl-users(a)lists.wfu.edu
> Subject: Re: [Gretl-users] Discrepancy between R and Gretl in Johansen
> cointegration
> Message-ID: <cac7e1ac-fa2e-f39c-afd5-60612c1e4853(a)gmx.net>
> Content-Type: text/plain; charset="utf-8"; Format="flowed"
>
> Am 31.01.2019 um 14:45 schrieb Reynaldo Senra:
>> As you can see and as I mentioned before, the test statistics is the
>> same, but the critical values (p-values) differ
>> Please, can you tell me why there is the discrepancy?
>>
> Hi,
> it sounds as if you're asking why the critical values are not the same
> as the p-values. Is that really the question? Because a critical value
> is a very different thing than a p-value.
>
> cheers,
> sven
> -------------- next part --------------
> An HTML attachment was scrubbed...
> URL: <http://lists.wfu.edu/pipermail/gretl-users/attachments/20190131/3b07ba71/...>
>
> ------------------------------
>
> Message: 2
> Date: Thu, 31 Jan 2019 16:07:51 +0100
> From: "Reynaldo Senra" <reynaldosenra(a)gmx.com>
> To: "Gretl list" <gretl-users(a)lists.wfu.edu>
> Subject: Re: [Gretl-users] Discrepancy between R and Gretl in Johansen
> cointegration
> Message-ID:
> <trinity-4562ac03-4ba2-4548-af9d-50bf3569ba99-1548947271415@msvc-mesg-gmx021>
>
> Content-Type: text/plain; charset="us-ascii"
>
> An HTML attachment was scrubbed...
> URL: <http://lists.wfu.edu/pipermail/gretl-users/attachments/20190131/e26ffcbf/...>
>
> ------------------------------
>
> _______________________________________________
> Gretl-users mailing list
> Gretl-users(a)lists.wfu.edu
> http://lists.wfu.edu/mailman/listinfo/gretl-users
>
> End of Gretl-users Digest, Vol 144, Issue 29
> ********************************************
--
________________________________________________________________________
Stefano Fachin
Professore Ordinario di Statistica Economica
Dip. di Scienze Statistiche
"Sapienza" Università di Roma
P.le A. Moro 5 - 00185 Roma - Italia
Tel. +39-06-49910834
fax +39-06-49910072
web http://stefanofachin.site.uniroma1.it/
5 years, 2 months
