APBRmetrics Forum Index APBRmetrics
The statistical revolution will not be televised.
 
 FAQFAQ   SearchSearch   MemberlistMemberlist   UsergroupsUsergroups   RegisterRegister 
 ProfileProfile   Log in to check your private messagesLog in to check your private messages   Log inLog in 

Basketball and MiniTab

 
Post new topic   Reply to topic    APBRmetrics Forum Index -> General discussion
View previous topic :: View next topic  
Author Message
mslan



Joined: 23 Apr 2007
Posts: 2

PostPosted: Thu Apr 26, 2007 8:51 am    Post subject: Basketball and MiniTab Reply with quote

Hey,

Whats going on everyone?

I was just wondering if anyone was familiar with a stats program called MiniTab and if anyone has any ideas if this can be helpful to me in building mutliple regressions to attempt to create my own predictions on basketball stats?

If anyone here has done anything relating to basketball within this program could you please lend a helping hand?

Thanks a lot.
Back to top
View user's profile Send private message
Ryan J. Parker



Joined: 23 Mar 2007
Posts: 88
Location: Charleston, SC

PostPosted: Thu Apr 26, 2007 9:53 am    Post subject: Reply with quote

I've never used it, so I can only suggest R (it's free).
Back to top
View user's profile Send private message Visit poster's website
tpryan



Joined: 11 Feb 2005
Posts: 64

PostPosted: Sat Apr 28, 2007 1:30 am    Post subject: Reply with quote

First of all, it is "Minitab". I have used it in teaching for almost 30 years and am a member of the Author Assistance Program of Minitab, Inc. since I have written statistics books that they list as Minitab Companion Texts.

It will do just about anything in multiple regression that you would want to do.

R, which was suggested, is best suited for specialized methods not available in the best known and most popular statistical software.

For routine work in multiple regression, Minitab is far superior.
Back to top
View user's profile Send private message
Ryan J. Parker



Joined: 23 Mar 2007
Posts: 88
Location: Charleston, SC

PostPosted: Sat Apr 28, 2007 6:53 am    Post subject: Reply with quote

Well I'd hope so... at $1,200 per single user license.
Back to top
View user's profile Send private message Visit poster's website
ChrisH



Joined: 02 Sep 2006
Posts: 3
Location: Chicago

PostPosted: Mon Apr 30, 2007 9:02 am    Post subject: Reply with quote

you can get a student copy for a lot less. or at least you could a couple years ago.
Back to top
View user's profile Send private message
AJax



Joined: 22 Jan 2006
Posts: 44
Location: The Prairies of the Great Middle West

PostPosted: Mon Apr 30, 2007 11:46 am    Post subject: Reply with quote

Minitab is $122 for students at my university. Stata is comparably priced, and better in my opinion. It's also cross-platform, and I think Minitab only works on windows. R is cool and open-source but it's more for the specialist.

A lot of things presented on this forum like figuring correlations can be done with spreadsheets. For the next bump up in complexity - least squares and logistic regression, model comparison - something like Minitab would be great; SPSS, Stata, and SAS would work just as well too.
Back to top
View user's profile Send private message Visit poster's website
tpryan



Joined: 11 Feb 2005
Posts: 64

PostPosted: Mon Apr 30, 2007 12:07 pm    Post subject: Reply with quote

Minitab has a 30-day free trial. Many of my Internet short course students have utilized the free trial as it fits the time frame for the course.
Back to top
View user's profile Send private message
jkubatko



Joined: 05 Jan 2005
Posts: 509
Location: Columbus, OH

PostPosted: Mon Apr 30, 2007 1:27 pm    Post subject: Reply with quote

tpryan wrote:
For routine work in multiple regression, Minitab is far superior.


Far superior? Or far easier? Big difference.
_________________
Regards,
Justin Kubatko
Basketball Stats!
Back to top
View user's profile Send private message Send e-mail Visit poster's website
tpryan



Joined: 11 Feb 2005
Posts: 64

PostPosted: Mon Apr 30, 2007 9:07 pm    Post subject: Reply with quote

jkubatko wrote:
tpryan wrote:
For routine work in multiple regression, Minitab is far superior.


Far superior? Or far easier? Big difference.

In my opinion, the primary drawback to R (and S-Plus) for the casual user is that it is necessary to write code to do simple things. As far as I'm concerned, that shouldn't be necessary.

R is inadequate for the casual user; Sandy Weisberg indicates problems with R for such users at http://www.stat.umn.edu/~sandy/UseR06.pdf.

I will undoubtedly use R eventually in working on the 2nd edition of my regression book, but only for the specialized stuff. (I have used S-Plus in the past for regression work.)

Minitab is not great for "modern" regression, but most users do not even use regression software capability that existed 20 years ago!

It is easy to write Minitab macros (I have written a few dozen) and the library of Minitab macros at Minitab, Inc. can be downloaded from there. Macros don't help on computer-intensive stuff, however, because the code in macros must be interpreted rather than compiled. So there is no hope of doing something like LTS (least trimmed sum of squares) efficiently in Minitab unless it is provided in a future release and Minitab, Inc. has not moved into areas such as robust regression.

So R and S-Plus easily outclass Minitab in certain areas of regression, but Minitab has good regression capabilities for the vast majority of users and uses.
Back to top
View user's profile Send private message
jkubatko



Joined: 05 Jan 2005
Posts: 509
Location: Columbus, OH

PostPosted: Mon Apr 30, 2007 10:00 pm    Post subject: Reply with quote

Thanks for the clarification, Tom. I think you have to be careful about using words like "far superior." Many people would mistake that as a statement about the accuracy of Minitab as compared to R.
_________________
Regards,
Justin Kubatko
Basketball Stats!
Back to top
View user's profile Send private message Send e-mail Visit poster's website
tpryan



Joined: 11 Feb 2005
Posts: 64

PostPosted: Tue May 01, 2007 1:41 pm    Post subject: Reply with quote

jkubatko wrote:
Thanks for the clarification, Tom. I think you have to be careful about using words like "far superior." Many people would mistake that as a statement about the accuracy of Minitab as compared to R.

Accuracy generally isn't a problem for most statistical procedures with today's algorithms for virtually any statistical software product. Accuracy was a problem decades ago, however, and datasets such as the Longley (1967) dataset were used to test the accuracy of regression output.
Back to top
View user's profile Send private message
jkubatko



Joined: 05 Jan 2005
Posts: 509
Location: Columbus, OH

PostPosted: Tue May 01, 2007 3:03 pm    Post subject: Reply with quote

tpryan wrote:
Accuracy generally isn't a problem for most statistical procedures with today's algorithms for virtually any statistical software product. Accuracy was a problem decades ago, however, and datasets such as the Longley (1967) dataset were used to test the accuracy of regression output.


I know that and of course you know that, but I doubt the general public does.
_________________
Regards,
Justin Kubatko
Basketball Stats!
Back to top
View user's profile Send private message Send e-mail Visit poster's website
mslan



Joined: 23 Apr 2007
Posts: 2

PostPosted: Wed May 02, 2007 6:34 pm    Post subject: Reply with quote

thanks a lot for all your advice gentlemen.
Back to top
View user's profile Send private message
Display posts from previous:   
Post new topic   Reply to topic    APBRmetrics Forum Index -> General discussion All times are GMT - 5 Hours
Page 1 of 1

 
Jump to:  
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum


Powered by phpBB © 2001, 2005 phpBB Group