In article <[EMAIL PROTECTED]>,
Robert J. MacG. Dawson <[EMAIL PROTECTED]> wrote:


>Fahd wrote:

>> Hi,
>>   Does anyone know how would it be possible to convert discrete data
>> to
>> continous data for regression analysis? 


><FLAME level="gentle"> Please, please, please. Asking for help on data
>analysis without saying as much as you can about what the data are and
>where they are from and what you want to know about them is like going
>to the doctor and saying you have a friend who thinks he might have VD.
></FLAME>

>       So, where are these data from and...

>               -Robert Dawson

It is even worse.  Every statistical procedure has 
underlying probability assumptions, and it is clear
that anyone asking a question on how to convert 
data so it can be used for regression analysis does
not have any idea of what the assumptions are, nor
what the results would mean.

While it is very often done, it is improper for the
statistician to construct the model; if the user
cannot make the assumptions, the statistician is no
longer a consultant, but a coauthor.

With the current packages, it is often the case that
nobody has any idea of what anything means, but the
believers in the computer output can sure do a lot
of damage by using the ritual.

-- 
This address is for information only.  I do not claim that these views
are those of the Statistics Department or of Purdue University.
Herman Rubin, Deptartment of Statistics, Purdue University
[EMAIL PROTECTED]         Phone: (765)494-6054   FAX: (765)494-0558
.
.
=================================================================
Instructions for joining and leaving this list, remarks about the
problem of INAPPROPRIATE MESSAGES, and archives are available at:
.                  http://jse.stat.ncsu.edu/                    .
=================================================================

Reply via email to