January 2006 - Gretl-users - gretlml.univpm.it

Performing a t test of mean difference with factors

by Florent Bresson

I would like to perform a test of mean difference on a series for different levels of a factor (usually I'm using R, but I decided to choose gretl for my students). Please, what's the command ? Florent Bresson ___________________________________________________________________________ Nouveau : téléphonez moins cher avec Yahoo! Messenger ! Découvez les tarifs exceptionnels pour appeler la France et l'international. Téléchargez sur http://fr.messenger.yahoo.com

18 years, 3 months

3
2
0 / 0

Variable name

by john w

Hi, The Variable name on the main Gretl window is limited up to 8 characters. This somethimes causes problems with duplicate variable names. For example you can have diffs, logs, sesonal adjustment and variable name like dL_VARNAME_sa. This is already over the limit. The problem occurs when importing data from other programs. Variable names are cut. Is it possible to expand, in one of the next versions, the variable names from 8 to 16 characters? There will still be enough place for Descriptive label. Thank you. Keep with good work! _________________________________________________________________ Are you using the latest version of MSN Messenger? Download MSN Messenger 7.5 today! http://messenger.msn.co.uk

18 years, 3 months

4
5
0 / 0

bug (?): double quote in script comment causes syntax error

by Sven Schreiber

Hi, just stumbled across the following: Syntax-Fehler in genr-Formel Error executing script: halting > scalar r = -0.2 # Korrelationskoeffizient ("rho" ist in gretl tabu) If I change the " to ' or omit them, everything is fine, so it's not serious. cheers, sven

18 years, 3 months

2
1
0 / 0

2 ISSUES: ARIMA and Windows metafiles

by John Paravantis PhD

Allin, Two suggestions for the next version of gretl, if you deem them worthy of attention: 1. The new forecasting platform of gretl is SUPERB and allows one to carry out model estimation and forecasting without the need for any additional calculations or backtransformations. Would you consider adding similar capabilities in ARMA, i.e. making the ARMA platform look like an ARIMA platform that would, in other words, allow the user to specify the degree of differencing and let gretl take care of the rest? 2. When gretl plots a scatterplot in which zero is included in the Y range, the zero reference line is plotted nicely as a DOTTED line and this translates correctly to either a PDF or an EPS graphics file. UNFORTUNATELY when copying/pasting or saving the file as a Windows metafile, this dotted line becomes SOLID. I am sure there is a solution to this in gnuplot BUT it should not be trivial. Would you consider fixing this so one may readily copy/paste graphic files into, say, a Word document? Regards John Paravantis University of Piraeus

18 years, 3 months

2
1
0 / 0

VAR roots plot

by Allin Cottrell

Sorry, can't remember who mentioned this, but someone said it would be nice to see the numerical values of the roots in this plot. This is done in CVS and will be in the next release: when you mouse over the "dot" representing a root you get a label showing the numerical value. Allin Cottrell

18 years, 3 months

4
14
0 / 0

Variable name and future requests

by john w

Thank you very much for your answer. I will be very greatefull if you expand variable name box. I'm using Gretl from Wake Forrest University and I must say that plenty of things are made in last years. Good work guys. But I also have some new suggestions for you...new features requests: 1. Data interpolation but from lower to high frequency data. Example from quarterlly to monthly. 2. The possibility to include exogenous variables in VAR/VECM. 3.Var inverse roots graph but followed with textual pop-up menu. It's difficult to see if some value is on or out the unit circle if it is close to the line. 4.FEVD followed by graph. 5. More stability tests for VAR/VECM (Cusum, breakpoint, 1-step Chow test, etc.) 6. F-tests of zero restrictions (Wald type Granger causality) for VECM. This are only the suggestions. I know that you have plenty of work and unfortunatelly I'm not a programer so I can't help you. Your comments are welcome! >On Thu, 19 Jan 2006, john w wrote: > >>Is it possible to expand, in one of the next versions, the variable names >>from 8 to 16 characters? > >I agree. In fact, I'd make it 32 characters. I can't see any technical >impediment, except going through the code to double-check everything will >be tedious and error-prone. > >Allin, your opinion? If you agree, I can start working on this in the next >days. I think this is basically a good idea (though I'd tend to go for 16 characters, not 32, for regular variable names). The main problem is with the various output-printing functions. For both console output and the output that goes to gui windows, everything is in a fixed format. Many of the output functions would have to be redesigned to avoid stuff breaking over lines given a wider variable name field. One possible redesign that would save width on numeric output fields would be to print all numbers such as coefficients and standard errors in a fixed width, as opposed to gretl's current practice of using decimal-aligned numbers printed to a given number of significant figures, using scientific notation only when absolutely needed. But I think gretl's current numeric output is nice and easy to read -- you can see the relative magnitude of coefficients at a glance. That said, I don't claim that gretl's current regression output is perfect and I'm certainly willing to consider proposals for redesign, within an 80-column format. Allin Cottrell -------------------------------------------------------------------------------- Previous message: [Gretl-users] Variable name _________________________________________________________________ Are you using the latest version of MSN Messenger? Download MSN Messenger 7.5 today! http://messenger.msn.co.uk

18 years, 3 months

5
9
0 / 0

Re: Gretl-users Digest, Vol 21, Issue 15

by John Paravantis

I should like to add my two pennies and say that expanding the 8 characters limit in variable names is an OUTSTANDING and long overdue suggestion! Even SPSS has gone to practically unlimited variable names in its latest versions. John Paravantis University of Piraeus >The Variable name on the main Gretl window is limited up to 8 characters. >This somethimes causes problems with duplicate variable names. >For example you can have diffs, logs, sesonal adjustment and variable name >like dL_VARNAME_sa. This is already over the limit. The problem occurs when >importing data from other programs. Variable names are cut. > >Is it possible to expand, in one of the next versions, the variable names >from 8 to 16 characters? >There will still be enough place for Descriptive label. > >Thank you. > > >I agree. In fact, I'd make it 32 characters. I can't see any technical >impediment, except going through the code to double-check everything will >be tedious and error-prone. > >Allin, your opinion? If you agree, I can start working on this in the >next days. > > >Riccardo `Jack' Lucchetti >Dipartimento di Economia >Università Politecnica delle Marche > >jack(a)dea.unian.it >http://www.econ.univpm.it/lucchetti > >

18 years, 3 months

1
0
0 / 0

gmm

by Arnaud Bervas

Sure, but my question was rather for more general estimators than OLS or even 2SLS with instrumental variables. I didn't intend to run a dynamic panel regression with Arellano and Bond estimator though. Actually, I was wondering if it would be possible to estimate Euler equations derived from a CCAPM model for instance.I guess it's not possible to program the implied numerical optimization with Gretl in its current state.

18 years, 3 months

4
3
0 / 0

gmm

by arnaud.bervas＠banque-france.fr

Sure, but my question was rather for more general estimators than OLS or even 2SLS with instrumental variables. I didn't intend to run a dynamic panel regression with Arellano and Bond estimator though. Actually, I was wondering if it would be possible to estimate Euler equations derived from a CCAPM model for instance.I guess it's not possible to program the implied numerical optimization with Gretl in its current state. ************************************************************** Ce mel est a l'attention exclusive des destinataires designes. Il peut contenir des informations confidentielles. Si vous le recevez par erreur, merci d'en informer sans delai l'expediteur. Le contenu de ce mel ne pourrait engager la responsabilite de la Banque de France que s'il a ete emis par une personne dument habilitee agissant dans le strict cadre des fonctions auxquelles elle est employee et a des fins non etrangeres a ses attributions. *************************************************************

18 years, 3 months

1
0
0 / 0

Re: Gretl-users Digest, Vol 21, Issue 9

by John Paravantis

BAD MOVE. I'd much rather have them readily apparent on the main screen. How about offering the user an OPTION? John Paravantis University of Piraeus Greece >>I had to re-install Gretl, but surprisingly, when I try to add a >>lagged variable to the dataset, it's no longer displayed on the >>main screen (as it used to be in my previous version). It appears >>in the window "Model" for regressions though. There's no such a >>problem for the first difference operator for intance. Does anyone >>know how to get lagged variables visible just like the others ? >> >> > >It seemed to me that displaying the names of lagged variables in the >main window just cluttered up the screen for no good reason, so I >made them "invisible" in current gretl. > >But if people think that was a bad move, I can reinstate them. > >Any other thoughts on this? > >Allin Cottrell > >

18 years, 3 months

5
6
0 / 0

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

Gretl-users January 2006