You are not logged in.

 Discussion: All Topics in Algebra II Topic: Curve of Best Fit for Car's Skidding Distance

 Post a new topic to the All Content in Algebra II discussion
 << see all messages in this topic < previous message | next message >

 Subject: RE: Curve of Best Fit for Car's Skidding Distance Author: Craig Date: Jan 9 2008
I have read some of the posts for this message--some give power function
regressions, one mentioned an exponential regression, another says "the
solutions is" and gives a quadratic function.

I believe, from a physics standpoint, that Newton would agree with a quadratic
function.  However, what is the source of the data?  Is it actual measurements,
or was it generated by some formula?  (my guess: it was generated by a formula
based on regressing on real data, but I have no way of knowing).

From a statistical standpoint, the small number of data points given would not
rule out any of the options presented, and indeed, a few others might work, as
well.  A glance at the residual plot for y = x^2 / 21 is less than ideal, even
though the maximum absolute error is less than 0.5.

From a technology standpoint, I saw mention of a graphing calculator (TI-89,
but most have regression capability these days) and a Shodor applet.  Many
spreadsheets also have regression capabilities; Excel, for instance, allows a
power regression but also allows a forced intercept of 0 for some (such as
polynomial).

I like this data set, and it can engender good conversation in my Statistics
class (next year, when we do the regressions unit, or later this year as we
review for the AP exam).  So whether your posting to Math Tools was intentional
or mistaken, I appreciate it!