This preview shows page 1. Sign up to view the full content.
Unformatted text preview: en close to linear.
(c) The parabola for the Saturday data is y = 0.003x2 − 0.21x +238.30. We have R2 = .47497
which isn’t good. Thus the data isn’t modeled well by a quadratic function, either.
(d) The Thursday linear model had my weight on January 1, 2010 at 193.77 pounds. The
Saturday models give 235.69 and 563.31 pounds, respectively. The Thursday line has
my weight going below 0 pounds in about ﬁve and a half years, so that’s no good. The
quadratic has a positive leading coeﬃcient which would mean unbounded weight gain
for the rest of my life. The Saturday line, which mathematically does not ﬁt the data at
all, yields a plausible weight prediction in the end. I think this is why grown-ups talk
about “Lies, Damned Lies and Statistics.”
4. (a) The quadratic model for the cats in Portage county is y = 1917803.54x2 − 16036408.29x +
24094857.7. Although R2 = .70888 this is not a good model because it’s so far oﬀ for
small values of x. Case in point, the model gives us 24,094,858 cats when x...
View Full Document