minitab stores residulas easily if you want it to ... either in regression
... or, you can do a simple regression with line fitted and, option to save
residuals ...
and you can do lots more ... anything you can get into a column ... or a
constant ... or matrix ... you can then save the worksheet containing those
values
for example ... say you have 3 predictors in c1 c2 c3 ... and a criterion
in c4 ... as a command you could:
mtb> regr c4 3 c1-c3;
subc> resi c5.
c5 will have the residuals ...
At 12:48 PM 6/8/00 -0500, Ken K. wrote:
>Its really a tough choice. They are both powerful packages.
>
>Both packages have gone well beyond your current Minitab 10. I also have
>to preface that I only have JMP 3.2 - I have not seen JMP 4 which has
>just been released.
>
>Both can do multiple regression about equally well.
>
>Minitab 13 has pretty easy to use checkbox options to create...
>-a histogram of residuals
>-a normal probability plot of residuals
>-a plot of residuals vs fit (y-hat)
>-a plot of residuals vs run order (order in worksheet)
>-a plot of residuals vs a specified variable
>Minitab also allows you to save the residuals to do other plotting.
>
>JMP 3 does an easy to obtain plot of residuals vs predicted (y-hat).
>Beyond that I think you need to save the residuals and do your own
>plots. JMP does allow you to save the residuals.
>
>By "generate estimates" I'll assume you mean that it calculates
>predicted values and the related error and confidence/prediction
>intervals....
>
>Minitab has an Options dialog box in which you can enter either
>individual values for x1, x2, ... or you can enter variable names that
>contain a list of multiple x1's, x2's, etc. The output includes the fit
>(y-hat), SE of fit, and the confidence & prediction intervals (you can
>select the confidence level).
>
>In JMP you can save predicted values (probably includes any obs. w/
>missing actual), SE of predicted, SE of individual, and confidence
>(Prediction Conf.) and prediction (Indiv. Conf.) intervals. Pretty much
>the same thing.
>
>For exploratory data analysis they are pretty close. Its pretty much a
>toss up.
>
>One other thing if it makes a difference, JMP has a nonlinear fitting
>tool; Minitab does not.
>
>My own thoughts (I've provided both packages to engineers in the past)
>in general is that Minitab is clearly easier to use for the occasional
>user (not my opinion but that of my engineers). To get the most out of
>JMP you need to be a fairly regular user, since many of its tools are
>somewhat hidden.
>
>[EMAIL PROTECTED] wrote:
>
>> Time for me to upgrade from Minitab 10.2 to the
>> new millenium. The current versions of both
>> Minitab and JMP have been recommended to me. The
>> analyses I primarily do are:
>> 1) developing predictive models for a fishery
>> management agency using multiple regression (2-3
>> independent variables),
>> 2) examination of the residuals from such models,
>> 3) generate estimates from the models, and
>> 4) general exploratory data analyses.
>>
>> One feature in particular that I would like to
>> see are built-in options to plot the predicted
>> and actual Y vs. the actual Y, and the residuals
>> against the various independent variables and Y.
>>
>> I would also be interested in software that has
>> Bayesian tools.
>>
>> So, which one - Minitab or JMP? Comments?
>> Suggestions?
>>
>> Sent via Deja.com http://www.deja.com/
>> Before you buy.
>
>
>
>===========================================================================
>This list is open to everyone. Occasionally, less thoughtful
>people send inappropriate messages. Please DO NOT COMPLAIN TO
>THE POSTMASTER about these messages because the postmaster has no
>way of controlling them, and excessive complaints will result in
>termination of the list.
>
>For information about this list, including information about the
>problem of inappropriate messages and information about how to
>unsubscribe, please see the web page at
>http://jse.stat.ncsu.edu/
>===========================================================================
==============================================================
dennis roberts, penn state university
educational psychology, 8148632401
http://roberts.ed.psu.edu/users/droberts/droberts.htm
===========================================================================
This list is open to everyone. Occasionally, less thoughtful
people send inappropriate messages. Please DO NOT COMPLAIN TO
THE POSTMASTER about these messages because the postmaster has no
way of controlling them, and excessive complaints will result in
termination of the list.
For information about this list, including information about the
problem of inappropriate messages and information about how to
unsubscribe, please see the web page at
http://jse.stat.ncsu.edu/
===========================================================================
