Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: riddle


From   Michael Hanson <mshanson@mac.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: riddle
Date   Wed, 19 Dec 2007 08:51:59 -0500

On Dec 19, 2007, at 8:03 AM, Maarten buis wrote:

In a previous post I claimed that the x-s in the graph below represent
the regression line (of x1 on x2, though I did not specify that
explicitly). Assuming that each symbol represent a datapoint, than this
is a mistake. It is a minor point, and does not change the point I made
in that previous post.

Does anybody else see my mistake? (maybe a nice exam question?)

| *x*
| **x**
| **x**
x1| **x**
| **x**
|**x**
|*x*
---------------
x2
Either your editor has rotated (and flipped?) your figure, or you have developed a "new technique" that minimizes the *horizontal* distance between observations and the "regression line". (I suspect some of us might call that regression of x2 on x1. I'm not so sure I'd characterize the difference as minor, however, especially in an introductory course: it cuts to the core of what a regression line represents.)

Agreed, a good ex*m question!

-- Mike


*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/




© Copyright 1996–2021 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index