A random sample of 200 observations of the continuous random variable X was taken. The values are summarised in the following table.
interval
1≤x<1.5
1.5≤x<2
2≤x<2.5
2.5≤x<3
3≤x<3.5
3.5≤x<4
x≥4
observed frequency
97
35
23
15
10
6
14
It is required to test the goodness of fit of the distribution with probability density function f given by
f(x)=⎩⎨⎧x32,0,x≥1,otherwise.
Most of the relevant expected frequencies, correct to 2 decimal places, are given in the following table.
interval
1≤x<1.5
1.5≤x<2
2≤x<2.5
2.5≤x<3
3≤x<3.5
3.5≤x<4
x≥4
expected frequency
111.11
a
18.00
9.78
b
3.83
12.50
(a) Show that a=38.89 and find the value of b correct to 2 decimal places.
(3)
(b) Carry out a goodness of fit test, at the 10% significance level, to test whether f is a satisfactory model for the data.
(6)
题目中文翻译
对连续随机变量 X 随机抽取 200 个观测值,结果汇总如下表。
区间
1≤x<1.5
1.5≤x<2
2≤x<2.5
2.5≤x<3
3≤x<3.5
3.5≤x<4
x≥4
观测频数
97
35
23
15
10
6
14
现需检验以下概率密度函数 f 所对应分布的拟合优度:
f(x)=⎩⎨⎧x32,0,x≥1,其他情况。
大部分相应的期望频数如下表所示,均精确到小数点后 2 位。
区间
1≤x<1.5
1.5≤x<2
2≤x<2.5
2.5≤x<3
3≤x<3.5
3.5≤x<4
x≥4
期望频数
111.11
a
18.00
9.78
b
3.83
12.50
(a) 证明 a=38.89,并求 b 的值,精确到小数点后 2 位。
(3)
(b) 在 10% 的显著性水平下进行拟合优度检验,检验 f 是否是这些数据的合适模型。
(6)
原卷通用作答要求
Answer all questions. You should use a calculator where appropriate.
You must show all necessary working clearly; no marks will be given for unsupported answers from a calculator.
Give non-exact numerical answers correct to 3 significant figures, or 1 decimal place for angles in degrees, unless a different level of accuracy is specified in the question.
The difference from the integrated value 5.90 arises from rounding the other expected frequencies.
(b)
解法一
思路
展开
把期望频数小于 5 的 [3.5,4) 与前一格合并。密度参数已知,按合并后格数确定自由度。
答题过程
展开
Test H0 that the stated density fits the observations against H1 that it does not. Combine [3,3.5) with [3.5,4). The combined observed count is 16, and the expected count is 200(3−2−4−2)=9.72222…. All six expected frequencies then exceed 5. Using unrounded frequencies,
There are 6−1=5 degrees of freedom. The upper 10% critical value is 9.236. Since the statistic exceeds it, we reject H0. There is sufficient evidence at the 10% level that the density is unsuitable for the data.
解法二
思路
展开
也可以将小频数格并入右侧的尾格。重新计算统计量后用相同的自由度判断。
答题过程
展开
Combining [3.5,4) with [4,∞) gives observed count 20 and expected count 200(3.5−2)=16.32653…. The other five groups are unchanged. In particular, [3,3.5) contributes (10−b)2/b with unrounded b. Summing the six contributions now gives χ2=10.04279…. The degrees of freedom are still 5. Since 10.04279…>9.236, we again reject the proposed model at the 10% level.