Math 6010, Fall 2004: Homework
Homework 1 Consider the following hypothetical data; it is regarding
a particular bird species:1
Wing length (cm) Age (days)
1.5 4.0
2.2 5.0
3.1 8.0
3.2 9.0
3.2 10.0
3.9 11.0
4.1 12.0
4.7 14.0
5.2 16.0
Use simple linear regression to predict the age of a newly-caught bird
whose wing is 4.0 centimeter long. Assuming normal errors, test the
hypothesis (95%) that wing length and age are linearly related.
Solution. Let the ages be denoted by y1, . . . , y9, and the wing-lengths
by x1, . . . , x9. You should compute:
¯x≈3.46 sx≈1.17
¯y≈9.89 sy≈3.92 Corr(x, y)≈0.99.
Now recall that the regression line is described by
y= ¯y+b
β(x−¯x),
where b
β= Corr(x, y)sy
sx
≈3.3.
So the regression line is (approximately) given by
y= 9.89 + 3.3(x−3.46).
Plug in x= 4 to obtain the regression estimate:
y= 9.89 + 3.3(4 −3.46) ≈11.21.
This answers the first question.
To answer the second question recall that
b
β∼Nβ, σ2
ns2
X.
1Borrowed from http://math.hws.edu/javamath/ryan/Regression.html
1
