题目
Problem
Stuart is investigating the relationship between Gross Domestic Product (GDP) and the size of the population for a particular country.
He takes a random sample of 9 9 9 years and records the size of the population, t t t millions, and the GDP, g g g billion dollars for each of these years.
The data are summarised as
n = 9 ∑ t = 7.87 ∑ g = 144.84 ∑ g 2 = 3624.41 n=9\qquad
\sum t=7.87\qquad
\sum g=144.84\qquad
\sum g^2=3624.41 n = 9 ∑ t = 7.87 ∑ g = 144.84 ∑ g 2 = 3624.41
S t t = 1.29 S t g = 40.25 S_{tt}=1.29\qquad S_{tg}=40.25 S tt = 1.29 S t g = 40.25
(a) Calculate the product moment correlation coefficient between t t t and g g g
(3)
(b) Give an interpretation of your product moment correlation coefficient.
(1)
(c) Find the equation of the least squares regression line of g g g on t t t in the form g = a + b t g=a+bt g = a + b t
(4)
(d) Give an interpretation of the value of b b b in your regression line.
(1)
(e) (i) Use the regression line from part (c) to estimate the GDP, in billions of dollars, for a population of 7000000 7000000 7000000
(2)
(ii) Comment on the reliability of your answer in part (i). Give a reason, in context, for your answer.
(1)
Using the regression line from part (c), Stuart estimates that for a population increase of x x x million there will be an increase of 0.1 0.1 0.1 billion dollars in GDP.
(f) Find the value of x x x
(2)
题目中文翻译
Stuart 正在研究某个国家的国内生产总值(GDP)与人口规模之间的关系。他随机抽取 9 9 9 个年份,并记录每一年的总人口 t t t (单位:百万)及 GDP g g g (单位:十亿美元)。
数据汇总如下:
n = 9 ∑ t = 7.87 ∑ g = 144.84 ∑ g 2 = 3624.41 n=9\qquad
\sum t=7.87\qquad
\sum g=144.84\qquad
\sum g^2=3624.41 n = 9 ∑ t = 7.87 ∑ g = 144.84 ∑ g 2 = 3624.41
S t t = 1.29 S t g = 40.25 S_{tt}=1.29\qquad S_{tg}=40.25 S tt = 1.29 S t g = 40.25
(a) 计算 t t t 与 g g g 的积矩相关系数。
(b) 解释你的积矩相关系数。
(c) 求 g g g 对 t t t 的最小二乘回归直线方程,并写成 g = a + b t g=a+bt g = a + b t 的形式。
(d) 解释回归直线中 b b b 的含义。
(e) (i) 使用 (c) 中的回归直线,估计人口为 7000000 7000000 7000000 时的 GDP,单位为十亿美元。
(ii) 评论 (i) 中估计值的可靠性,并结合题目情境说明理由。
Stuart 使用 (c) 中的回归直线估计:人口每增加 x x x 百万,GDP 将增加 0.1 0.1 0.1 十亿美元。
(f) 求 x x x 的值。
解答
(a)
解法一
思路
展开
先求 S g g S_{gg} S g g ,再代入 PMCC 公式。
答题过程
展开
S g g = ∑ g 2 − ( ∑ g ) 2 n = 3624.41 − 144.84 2 9 = 1293.4516 … \begin{align*}
S_{gg}
=&\,\sum g^2-\frac{(\sum g)^2}{n}\\
=&\,3624.41-\frac{144.84^2}{9}\\
=&\,1293.4516\ldots
\end{align*} S g g = = = ∑ g 2 − n ( ∑ g ) 2 3624.41 − 9 144.8 4 2 1293.4516 …
Therefore
r = S t g S t t S g g = 40.25 1.29 ( 1293.4516 … ) = 0.985 … \begin{align*}
r
=&\,\frac{S_{tg}}{\sqrt{S_{tt}S_{gg}}}\\
=&\,\frac{40.25}{\sqrt{1.29(1293.4516\ldots)}}\\
=&\,0.985\ldots
\end{align*} r = = = S tt S g g S t g 1.29 ( 1293.4516 … ) 40.25 0.985 …
So
r = 0.985. \begin{align*}
r=0.985.
\end{align*} r = 0.985.
(b)
解法一
思路
展开
r r r 为正且很接近 1 1 1 ,表示 population 越大,GDP 一般越高。
答题过程
展开
As the population increases, the GDP generally increases.
(c)
解法一
思路
展开
回归线 g g g on t t t 的斜率是 S t g / S t t S_{tg}/S_{tt} S t g / S tt ,截距是 g ˉ − b t ˉ \bar{g}-b\bar{t} g ˉ − b t ˉ 。
答题过程
展开
b = S t g S t t = 40.25 1.29 = 31.20155 … \begin{align*}
b
=&\,\frac{S_{tg}}{S_{tt}}\\
=&\,\frac{40.25}{1.29}\\
=&\,31.20155\ldots
\end{align*} b = = = S tt S t g 1.29 40.25 31.20155 …
Also,
a = g ˉ − b t ˉ = 144.84 9 − 31.20155 … ( 7.87 9 ) = − 11.19068 … \begin{align*}
a
=&\,\bar{g}-b\bar{t}\\
=&\,\frac{144.84}{9}-31.20155\ldots\left(\frac{7.87}{9}\right)\\
=&\,-11.19068\ldots
\end{align*} a = = = g ˉ − b t ˉ 9 144.84 − 31.20155 … ( 9 7.87 ) − 11.19068 …
Thus
g = − 11.2 + 31.2 t . \begin{align*}
g=-11.2+31.2t.
\end{align*} g = − 11.2 + 31.2 t .
(d)
解法一
思路
展开
t t t 的单位是 million,g g g 的单位是 billion dollars。所以斜率表示 population 每增加 1 1 1 million,GDP 平均增加多少 billion dollars。
答题过程
展开
For each increase of 1 1 1 million in population, GDP increases by about 31.2 31.2 31.2 billion dollars on average.
(e)(i)
解法一
思路
展开
7000000 7000000 7000000 人即 7 7 7 million,所以代入 t = 7 t=7 t = 7 。
答题过程
展开
g = − 11.2 + 31.2 ( 7 ) = 207.2. \begin{align*}
g
=&\,-11.2+31.2(7)\\
=&\,207.2.
\end{align*} g = = − 11.2 + 31.2 ( 7 ) 207.2.
The estimated GDP is
207 billion dollars \begin{align*}
207\text{ billion dollars}
\end{align*} 207 billion dollars
to 3 3 3 significant figures.
(e)(ii)
解法一
思路
展开
样本的平均 population 只有 7.87 9 ≈ 0.874 \dfrac{7.87}{9}\approx0.874 9 7.87 ≈ 0.874 million,而 7 7 7 million 远大于这个范围,估计不可靠。
答题过程
展开
The estimate is unreliable because a population of 7 7 7 million is much greater than the mean population in the sample.
(f)
解法一
思路
展开
GDP 的增加量等于斜率乘以 population 增加量。
答题过程
展开
0.1 = 31.2 x . \begin{align*}
0.1=31.2x.
\end{align*} 0.1 = 31.2 x .
So
x = 0.1 31.2 = 0.003205 … \begin{align*}
x=\frac{0.1}{31.2}=0.003205\ldots
\end{align*} x = 31.2 0.1 = 0.003205 …
Therefore
x = 0.00321 \begin{align*}
x=0.00321
\end{align*} x = 0.00321
million people.
解法二
思路
展开
在同一条回归线上选取两个 GDP 相差 0.1 0.1 0.1 的点,两式相减后截距会消去,人口差就是 x x x 。
答题过程
展开
Let the two population values be t t t and t + x t+x t + x . From the regression line,
g ( t + x ) − g ( t ) = [ − 11.2 + 31.2 ( t + x ) ] − ( − 11.2 + 31.2 t ) = 31.2 x . \begin{align*}
g(t+x)-g(t)
=&\,\big[-11.2+31.2(t+x)\big]\\
&\,-\big(-11.2+31.2t\big)\\
=&\,31.2x.
\end{align*} g ( t + x ) − g ( t ) = = [ − 11.2 + 31.2 ( t + x ) ] − ( − 11.2 + 31.2 t ) 31.2 x .
The increase in GDP is 0.1 0.1 0.1 , so
31.2 x = 0.1 , x = 0.003205 … \begin{align*}
31.2x=&\,0.1,\\
x=&\,0.003205\ldots
\end{align*} 31.2 x = x = 0.1 , 0.003205 …
Hence
x = 0.00321 \begin{align*}
x=0.00321
\end{align*} x = 0.00321
million people.