题目
Problem
A biologist records the length, y cm, and the weight, w kg, of 50 rabbits.
The following summary statistics are obtained.
∑y=2015,∑y2=81938.5,∑w=125,
Sww=72.25,Syw=219.55.
(a) (i) Show that Syy=734.
(ii) Calculate the product moment correlation coefficient for these data.
Give your answer to 3 decimal places.
(3)
(b) Interpret the product moment correlation coefficient in context.
(1)
(c) State, giving a reason, whether the value of the product moment correlation coefficient is consistent with the use of a linear regression model.
(1)
(d) Find the equation of the regression line of w on y, giving your answer in the form w=a+by.
(4)
A rabbit named Jeff has length 45 cm.
(e) Use your regression equation to estimate the weight of Jeff.
(2)
解答
(a)(i)
解法一
思路
展开
用
Syy=∑y2−n(∑y)2.
题目要求 show that,所以要写出代入式和计算结果。
答题过程
展开
Syy====∑y2−n(∑y)281938.5−502015281938.5−81204.5734.
(a)(ii)
解法一
思路
展开
PMCC 的公式是
r=SyySwwSyw.
答题过程
展开
r===SyySwwSyw734(72.25)219.550.95338….
So, to 3 decimal places,
r=0.953.
(b)
解法一
思路
展开
r 接近 1,表示长度和重量之间有强正相关。解释时要放回语境:兔子越长,通常越重。
答题过程
展开
There is a strong positive correlation between length and weight.
In general, the longer the rabbit, the greater its weight.
(c)
解法一
思路
展开
r=0.953 很接近 1,说明数据点应当相当接近一条正斜率直线,所以适合线性回归模型。
答题过程
展开
Yes. The value of r is close to 1, so the data are consistent with a linear regression model.
(d)
解法一
思路
展开
回归线 w=a+by 中,
b=SyySyw.
然后用均值点 (yˉ,wˉ) 在回归线上,求截距 a。
答题过程
展开
The gradient of the regression line of w on y is
b===SyySyw734219.550.29911….
Also,
yˉ=502015=40.3,wˉ=50125=2.5.
Using a=wˉ−byˉ,
a==2.5−(0.29911…)(40.3)−9.554….
Therefore the regression equation is
w=−9.55+0.299y.
(e)
解法一
思路
展开
把 Jeff 的长度 y=45 代入上一小题的回归方程。
答题过程
展开
When y=45,
w==−9.55+0.299(45)3.905.
The estimated weight of Jeff is
3.91 kg.