For two years I have been using GRETL while volunteering at my wife's high school helping the research design class with their
social and behavioral science fair projects. I have to say that I am impressed with the package so far as an open source package.
So, let me briefly list the strengths and weaknesses that I have seen so far with the package.
Striengths
1.) User friendly program which presents variable choice in such a way that it makes creating a model intuitive to theory. Much
like STATA menus, it asks for DV and IV's and does not depend on a variable list like R or the STATA programing language. When 9th
and 11th graders can use the program and build models to test theories, then you have a program that is excellent for teaching
beginning statistics.
2.) The program makes an interesting and useful tool for manipulating data for use in R programing. GRETL presents data in a way
that makes data manipulation and checking for errors in R data more intuitive.
3.) The ability to open R directly from GRETL makes the program an excellent gateway to the programing language. You can ease a
student into the more "hard-core" realm of R without scaring him or her away.
4.) GRETL allows for the importation of data from a large number of sources including STATA. When combined with the ability to
transfer data directly to R, GRETL may eventually give STAT/Transfer a run for its money.
5.) Open source means that as long as a C based programing community supports the program it will continue to grow and develop.
GRETL represents a good start for such a open source program.
Problems
1.) GRETL represents a good START. GRETL lacks many of the functions that are available in commercial statistical packages. While
over time this may be remedied, it still represents a drawback. Yes, this is reminded in part through the ease in which these
functions can be accessed through R, but as a stand alone package GRETL is still behind its competitors.
2.) While Time-Series is included including support for Cross-Sectional Time Series, the program lacks the functionality that those
who use RATS and other Time Series specific programs expect. While this could also be said for other packages as well, and is
hardly a fair critique it still should be pointed out.
Over all, I would say that the package is strongest as a tool for use in introducing statistics and R. However it needs further
development before it can be a serious contender with STATA or SPSS, the two programs that are most like GRETL in functionality.
TL Long
Full Time Instructor
Department of Liberal Arts
Temple College
[log in to unmask]
------- Original Message -------
From : Paul Gronke[mailto:[log in to unmask]]
Sent : 7/22/2008 3:37:02 PM
To : [log in to unmask]
Cc :
Subject : RE: [POLMETH] Gretl
I am working with a group that uses a statistical package called Gretl
(Gnu Regression Econometrics and Time-series, not sure where the L is
from!).
Any opinions or experiences with this package? it does say it reads
Stata .dta files directly which is important for us.
---
Paul Gronke Ph: 503-517-7393
Professor Fax: 734-661-0801
Reed College
3203 SE Woodstock Blvd.
Portland OR 97202
**********************************************************
Political Methodology E-Mail List
Editors: Melanie Goodrich, <[log in to unmask]>
Delia Bailey, <[log in to unmask]>
**********************************************************
Send messages to [log in to unmask]
To join the list, cancel your subscription, or modify
your subscription settings visit:
http://polmeth.wustl.edu/polmeth.php
**********************************************************
**********************************************************
Political Methodology E-Mail List
Editors: Melanie Goodrich, <[log in to unmask]>
Delia Bailey, <[log in to unmask]>
**********************************************************
Send messages to [log in to unmask]
To join the list, cancel your subscription, or modify
your subscription settings visit:
http://polmeth.wustl.edu/polmeth.php
**********************************************************
|
