题目
Problem
The production cost, £c c c million, of a film and the total ticket sales, £t t t million, earned by the film are recorded for a sample of 40 40 40 films.
Some summary statistics are given below.
∑ c = 1634 ∑ t = 1361 ∑ t 2 = 82873 \sum c=1634\qquad
\sum t=1361\qquad
\sum t^2=82873 ∑ c = 1634 ∑ t = 1361 ∑ t 2 = 82873
∑ c t = 83634 S c c = 28732.1 \sum ct=83634\qquad
S_{cc}=28732.1 ∑ c t = 83634 S cc = 28732.1
(a) Find the exact value of S t t S_{tt} S tt and the exact value of S c t S_{ct} S c t
(3)
(b) Calculate the value of the product moment correlation coefficient for these data.
(2)
(c) Give an interpretation of your answer to part (b)
(1)
(d) Show that the equation of the linear regression line of t t t on c c c can be written as
t = − 5.84 + 0.976 c t=-5.84+0.976c t = − 5.84 + 0.976 c
where the values of the intercept and gradient are given to 3 3 3 significant figures.
(3)
(e) Find the expected total ticket sales for a film with a production cost of £90 90 90 million.
(2)
Using the regression line in part (d)
(f) find the range of values of the production cost of a film for which the total ticket sales are less than 80 % 80\% 80% of its production cost.
(2)
题目中文翻译
对一个由 40 40 40 部电影组成的样本,记录每部电影的制作成本 £c c c million,以及其获得的总票房 £t t t million。题目给出了若干汇总统计量。
(a) 求 S t t S_{tt} S tt 和 S c t S_{ct} S c t 的准确值。
(b) 计算这些数据的积矩相关系数。
(c) 解释你在 (b) 中所得答案的含义。
(d) 证明 t t t 关于 c c c 的线性回归方程可写成
t = − 5.84 + 0.976 c , t=-5.84+0.976c, t = − 5.84 + 0.976 c ,
其中截距和斜率均保留 3 3 3 位有效数字。
(e) 求制作成本为 £90 90 90 million 的电影的预期总票房。
使用 (d) 中的回归直线:
(f) 求电影制作成本的取值范围,使总票房小于制作成本的 80 % 80\% 80% 。
解答
(a)
解法一
思路
展开
直接用 S S S 的定义式:平方和减去校正项。
答题过程
展开
S t t = ∑ t 2 − ( ∑ t ) 2 n = 82873 − 1361 2 40 = 36564.975. \begin{align*}
S_{tt}
=&\,\sum t^2-\frac{(\sum t)^2}{n}\\
=&\,82873-\frac{1361^2}{40}\\
=&\,36564.975.
\end{align*} S tt = = = ∑ t 2 − n ( ∑ t ) 2 82873 − 40 136 1 2 36564.975.
Also,
S c t = ∑ c t − ∑ c ∑ t n = 83634 − 1634 ( 1361 ) 40 = 28037.15. \begin{align*}
S_{ct}
=&\,\sum ct-\frac{\sum c\sum t}{n}\\
=&\,83634-\frac{1634(1361)}{40}\\
=&\,28037.15.
\end{align*} S c t = = = ∑ c t − n ∑ c ∑ t 83634 − 40 1634 ( 1361 ) 28037.15.
(b)
解法一
思路
展开
代入 PMCC 公式。
答题过程
展开
r = S c t S c c S t t = 28037.15 28732.1 ( 36564.975 ) = 0.865 … \begin{align*}
r
=&\,\frac{S_{ct}}{\sqrt{S_{cc}S_{tt}}}\\
=&\,\frac{28037.15}{\sqrt{28732.1(36564.975)}}\\
=&\,0.865\ldots
\end{align*} r = = = S cc S tt S c t 28732.1 ( 36564.975 ) 28037.15 0.865 …
So
r = 0.865 \begin{align*}
r=0.865
\end{align*} r = 0.865
to 3 3 3 significant figures.
(c)
解法一
思路
展开
r r r 是正且较接近 1 1 1 ,说明成本越高,票房销售额一般越高。
答题过程
展开
There is a positive correlation: films with higher production costs generally have higher total ticket sales.
(d)
解法一
思路
展开
回归线 t t t on c c c 的斜率是 S c t / S c c S_{ct}/S_{cc} S c t / S cc ,截距是 t ˉ − b c ˉ \bar{t}-b\bar{c} t ˉ − b c ˉ 。
答题过程
展开
b = S c t S c c = 28037.15 28732.1 = 0.97581 … \begin{align*}
b
=&\,\frac{S_{ct}}{S_{cc}}\\
=&\,\frac{28037.15}{28732.1}\\
=&\,0.97581\ldots
\end{align*} b = = = S cc S c t 28732.1 28037.15 0.97581 …
Also,
a = t ˉ − b c ˉ = 1361 40 − 0.97581 … ( 1634 40 ) = − 5.8369 … \begin{align*}
a
=&\,\bar{t}-b\bar{c}\\
=&\,\frac{1361}{40}-0.97581\ldots\left(\frac{1634}{40}\right)\\
=&\,-5.8369\ldots
\end{align*} a = = = t ˉ − b c ˉ 40 1361 − 0.97581 … ( 40 1634 ) − 5.8369 …
Therefore, to 3 3 3 significant figures,
t = − 5.84 + 0.976 c . \begin{align*}
t=-5.84+0.976c.
\end{align*} t = − 5.84 + 0.976 c .
(e)
解法一
思路
展开
把 c = 90 c=90 c = 90 代入回归线。单位仍是 million pounds。
答题过程
展开
t = − 5.84 + 0.976 ( 90 ) = 82.0. \begin{align*}
t
=&\,-5.84+0.976(90)\\
=&\,82.0.
\end{align*} t = = − 5.84 + 0.976 ( 90 ) 82.0.
The expected total ticket sales are approximately
£ 82 million . \begin{align*}
\text{£}82\text{ million}.
\end{align*} £ 82 million .
(f)
解法一
思路
展开
“ticket sales less than 80 % 80\% 80% of production cost” 即 t < 0.8 c t<0.8c t < 0.8 c ,再用回归线代入。
答题过程
展开
We need
t < 0.8 c . t<0.8c. t < 0.8 c .
Using the regression line,
− 5.84 + 0.976 c < 0.8 c 0.176 c < 5.84 c < 33.1818 … \begin{align*}
-5.84+0.976c&<0.8c\\
0.176c&<5.84\\
c&<33.1818\ldots
\end{align*} − 5.84 + 0.976 c 0.176 c c < 0.8 c < 5.84 < 33.1818 …
Therefore
c < 33.2. c<33.2. c < 33.2.
So the production cost is less than about £33.2 33.2 33.2 million.