113 年 國立清華大學科技管理研究所甲組《統計學》

📄 試題原卷 免費註冊後即可對照原始考卷 PDF免費註冊

第 1 題

  1. Although studies continue to show smoking leads to significant health problems, 25% of adults in the United States smoke. Consider a group of 400 adults. What is the probability that fewer than 60 smoke? (10%)

登入後即可作答並保存紀錄。

這一題的完整詳解

本題考驗二項分布的近似以及常態分布的應用。

一個簡單的二項分布問題,我們將使用常態近似來求解。

令 XX 為這 400 位成人中吸菸的人數。
已知吸菸的機率 p=0.25p = 0.25,樣本數 n=400n = 400。
XX 服從二項分布 B(n,p)B(n, p),即 X∼B(400,0.25)X \sim B(400, 0.25)。

二項分布的平均數為 μ=np=400×0.25=100\mu = np = 400 \times 0.25 = 100。
二項分布的變異數為 σ2=np(1−p)=400×0.25×(1−0.25)=400×0.25×0.75=100×0.75=75\sigma^2 = np(1-p) = 400 \times 0.25 \times (1-0.25) = 400 \times 0.25 \times 0.75 = 100 \times 0.75 = 75。
標準差為 σ=75=25×3=53≈5×1.732=8.66\sigma = \sqrt{75} = \sqrt{25 \times 3} = 5\sqrt{3} \approx 5 \times 1.732 = 8.66。

由於 nn 夠大,且 np=100>5np = 100 > 5 且 n(1−p)=400×0.75=300>5n(1-p) = 400 \times 0.75 = 300 > 5,我們可以利用常態分布來近似二項分布。

🔒

後續完整解題步驟與【答案】

免費註冊,享三天全站完整詳解閱覽。

免費註冊

第 2 題

  1. To analyze the risk, or volatility, associated with investing in AAA company common stock, a sample of the eight quarterly total returns was identified as shown below. The percent total return includes the stock price change plus the dividend payment for the quarter.
    18.0 -12.5 15.2 11.6 8.5 -2.8 -17.7 12.3
    Based on the above data, construct a 95% confidence interval for the population standard deviation. (20%)

登入後即可作答並保存紀錄。

這一題的完整詳解

本題考驗根據樣本資料建構母體標準差的信賴區間。

我們需要根據樣本資料,建構母體標準差 σ\sigma 的 95% 信賴區間。
當母體變異數 σ2\sigma^2 未知時,我們通常使用卡方分布 (Chi-squared distribution) 來建構 σ2\sigma^2 的信賴區間,進而推導出 σ\sigma 的信賴區間。

首先,計算樣本的平均數 xˉ\bar{x} 和樣本標準差 ss。
樣本數 n=8n=8。
樣本數據:18.0, -12.5, 15.2, 11.6, 8.5, -2.8, -17.7, 12.3

計算樣本平均數 xˉ\bar{x}:
xˉ=18.0−12.5+15.2+11.6+8.5−2.8−17.7+12.38\bar{x} = \frac{18.0 - 12.5 + 15.2 + 11.6 + 8.5 - 2.8 - 17.7 + 12.3}{8}
xˉ=44.68=5.575\bar{x} = \frac{44.6}{8} = 5.575

計算樣本變異數 s2s^2:
s2=∑i=1n(xi−xˉ)2n−1s^2 = \frac{\sum_{i=1}^{n}(x_i - \bar{x})^2}{n-1}
首先計算平方和 ∑i=1n(xi−xˉ)2\sum_{i=1}^{n}(x_i - \bar{x})^2:
(18.0−5.575)2=(12.425)2=154.380625(18.0 - 5.575)^2 = (12.425)^2 = 154.380625
(−12.5−5.575)2=(−18.075)2=326.705625(-12.5 - 5.575)^2 = (-18.075)^2 = 326.705625
(15.2−5.575)2=(9.625)2=92.640625(15.2 - 5.575)^2 = (9.625)^2 = 92.640625
(11.6−5.575)2=(6.025)2=36.300625(11.6 - 5.575)^2 = (6.025)^2 = 36.300625
(8.5−5.575)2=(2.925)2=8.555625(8.5 - 5.575)^2 = (2.925)^2 = 8.555625
(−2.8−5.575)2=(−8.375)2=70.140625(-2.8 - 5.575)^2 = (-8.375)^2 = 70.140625
(−17.7−5.575)2=(−23.275)2=541.725625(-17.7 - 5.575)^2 = (-23.275)^2 = 541.725625
(12.3−5.575)2=(6.725)2=45.225625(12.3 - 5.575)^2 = (6.725)^2 = 45.225625

🔒

後續完整解題步驟與【答案】

免費註冊,享三天全站完整詳解閱覽。

免費註冊

第 3 題

  1. Periodically, Merrill Lynch consultants are asked to evaluate Merrill Lynch financial consultants and services. Higher ratings on the client satisfaction survey indicate better service, with 7 the maximum rating. Independent samples of service ratings for two financial consultants are summarized here. Consultant A has 10 years of experience, whereas Consultant B has 1 year of experience. Here use a=.05a = .05 and test to see whether the consultant with more experience has the higher population mean service rating. (10%)
Consultant AConsultant B
n1=16n1 = 16n2=10n2 = 10
mean of x1 = 6.82mean of x2 = 6.25
s1 = 0.64s2 = 0.75

a. Use a=.05a = .05 and test to see whether the consultant with more experience has the higher population mean service rating. (10%)
b. Suppose X=1X=1 if the client from Consultant A; X=0X=0 if the client from Consultant B. YY is service rating score. Compute the correlation coefficient between XX and YY. (20%)

登入後即可作答並保存紀錄。

這一題的完整詳解

核心觀念

本題考查兩個觀念:

  1. 兩個獨立母體平均數差異的單尾檢定:判斷年資較深的顧問 A 是否具有較高的母體平均服務評分。
  2. 二元變數與連續變數的相關係數:當 XX 為 00、11 二元變數,YY 為連續評分時,可使用點二系列相關係數(point-biserial correlation)。

令:

  • 顧問 A:n1=16, xˉ1=6.82, s1=0.64n_1=16,\ \bar{x}_1=6.82,\ s_1=0.64
  • 顧問 B:n2=10, xˉ2=6.25, s2=0.75n_2=10,\ \bar{x}_2=6.25,\ s_2=0.75

樣本平均數差為:

xˉ1−xˉ2=6.82−6.25=0.57\bar{x}_1-\bar{x}_2=6.82-6.25=0.57

a. 檢定顧問 A 的母體平均評分是否較高

假設

令 μ1\mu_1 為顧問 A 的母體平均服務評分,μ2\mu_2 為顧問 B 的母體平均服務評分。

因題目要檢定顧問 A 是否較高,採右尾檢定:

H0:μ1−μ2≤0H_0:\mu_1-\mu_2\leq 0 H1:μ1−μ2>0H_1:\mu_1-\mu_2>0

顯著水準為:

α=0.05\alpha=0.05

由於題目未明確要求兩母體變異數相等,採用 Welch 兩獨立樣本 tt 檢定。

檢定統計量

Welch tt 統計量為:

t=(xˉ1−xˉ2)−0s12n1+s22n2t= \frac{(\bar{x}_1-\bar{x}_2)-0} {\sqrt{\frac{s_1^2}{n_1}+\frac{s_2^2}{n_2}}}

代入數值:

t=6.82−6.250.64216+0.75210t= \frac{6.82-6.25} {\sqrt{\frac{0.64^2}{16}+\frac{0.75^2}{10}}} =0.570.0256+0.05625=0.570.2861≈1.99= \frac{0.57} {\sqrt{0.0256+0.05625}} = \frac{0.57}{0.2861} \approx 1.99

Welch 自由度為:

ν=(s12n1+s22n2)2(s12/n1)2n1−1+(s22/n2)2n2−1≈16.94\nu= \frac{ \left(\frac{s_1^2}{n_1}+\frac{s_2^2}{n_2}\right)^2 }{ \frac{\left(s_1^2/n_1\right)^2}{n_1-1} + \frac{\left(s_2^2/n_2\right)^2}{n_2-1} } \approx 16.94

可取自由度 ν≈17\nu\approx 17。

臨界值與判定

右尾檢定、α=0.05\alpha=0.05、自由度約為 1717:

t0.05,17≈1.740t_{0.05,17}\approx 1.740

比較:

t=1.99>1.740t=1.99>1.740

因此拒絕 H0H_0。其單尾 pp 值約為:

p≈0.031<0.05p\approx 0.031<0.05

結論

在顯著水準 0.050.05 下,有充分統計證據支持:

μ1>μ2\mu_1>\mu_2

也就是說,年資較深的顧問 A,其母體平均服務評分顯著高於顧問 B。


b. 計算 XX 與 YY 的相關係數

方法一:點二系列相關係數

當 XX 是二元變數、YY 是連續變數時,相關係數可用:

rpb=Yˉ1−Yˉ0sYn1n0n2r_{pb} = \frac{\bar{Y}_1-\bar{Y}_0}{s_Y} \sqrt{\frac{n_1n_0}{n^2}}

其中:

🔒

後續完整解題步驟與【答案】

免費註冊,享三天全站完整詳解閱覽。

免費註冊

第 4 題

  1. AAA Businessweek subscriber asked, "in the past 12 months, when traveling for business, what type of airline ticket did you purchase most often?" A second question asked if the type of airline ticket purchased most often was for domestic or international.
Type of TicketDomesticInternational
First Class3025
Business Class90120
Economy Class520130

Using a 0.05 level of significance, is the type of ticket purchased independent of the type of flight? What is your conclusion? (20%)

登入後即可作答並保存紀錄。

這一題的完整詳解

本題考驗卡方獨立性檢定 (Chi-squared Test of Independence)。

我們需要檢定購買的機票類型(First Class, Business Class, Economy Class)是否與飛行類型(Domestic, International)獨立。

步驟 1:設定虛無假設與對立假設
H0H_0: 機票類型與飛行類型是獨立的。
H1H_1: 機票類型與飛行類型不是獨立的(即存在關聯)。

步驟 2:計算邊際總計和總計

Type of TicketDomestic (Col 1)International (Col 2)Row Total
First Class302555
Business Class90120210
Economy Class520130650
Column Total640275915 (N)

總樣本數 N=915N = 915。

步驟 3:計算期望次數 EijE_{ij}
在虛無假設(獨立性)下,期望次數的計算公式為:
Eij=(Row Totali)×(Column Totalj)NE_{ij} = \frac{(\text{Row Total}_i) \times (\text{Column Total}_j)}{N}

  • E11E_{11} (First Class, Domestic): 55×640915=35200915≈38.4699\frac{55 \times 640}{915} = \frac{35200}{915} \approx 38.4699
  • E12E_{12} (First Class, International): 55×275915=15125915≈16.5301\frac{55 \times 275}{915} = \frac{15125}{915} \approx 16.5301
  • E21E_{21} (Business Class, Domestic): 210×640915=134400915≈146.8852\frac{210 \times 640}{915} = \frac{134400}{915} \approx 146.8852
  • E22E_{22} (Business Class, International): 210×275915=57750915≈63.1148\frac{210 \times 275}{915} = \frac{57750}{915} \approx 63.1148
  • E31E_{31} (Economy Class, Domestic): 650×640915=416000915≈454.6448\frac{650 \times 640}{915} = \frac{416000}{915} \approx 454.6448
  • E32E_{32} (Economy Class, International): 650×275915=178750915≈195.3552\frac{650 \times 275}{915} = \frac{178750}{915} \approx 195.3552
🔒

後續完整解題步驟與【答案】

免費註冊,享三天全站完整詳解閱覽。

免費註冊

第 5 題

  1. To study the effect of temperature on yield in a chemical process, four batches were produced at each of three temperature levels. The results follow. Construct an analysis of variance table. Use a .05 level of significance to test whether the temperature level has an effect on the mean yield of the process. (20%)
Temperature50°C60°C70°C
343023
243529
343228
253625

登入後即可作答並保存紀錄。

這一題的完整詳解

本題考驗單因子變異數分析 (One-Way ANOVA)。

我們需要檢定不同的溫度水平(50°C, 60°C, 70°C)是否對化學製程的平均產量有顯著影響。

步驟 1:設定虛無假設與對立假設
令 μ1,μ2,μ3\mu_1, \mu_2, \mu_3 分別為溫度水平 50°C, 60°C, 70°C 下的母體平均產量。
H0:μ1=μ2=μ3H_0: \mu_1 = \mu_2 = \mu_3 (所有溫度水平下的平均產量都相同)
H1:至少有一個 μi 不同H_1: \text{至少有一個 } \mu_i \text{ 不同} (至少有一個溫度水平對平均產量有顯著影響)

步驟 2:整理數據並計算基本統計量
有 3 個處理組(溫度水平),每個組有 4 個觀察值。
總觀察數 N=3×4=12N = 3 \times 4 = 12。

Temperature50°C (i=1i=1)60°C (i=2i=2)70°C (i=3i=3)
343023
243529
343228
253625
Sum (TiT_i)34+24+34+25=11734+24+34+25 = 11730+35+32+36=13330+35+32+36 = 13323+29+28+25=10523+29+28+25 = 105
Mean (xˉi\bar{x}_i)117/4=29.25117/4 = 29.25133/4=33.25133/4 = 33.25105/4=26.25105/4 = 26.25
Sum of Squares (SSiSS_i)342+242+342+252=1156+576+1156+625=351334^2+24^2+34^2+25^2 = 1156+576+1156+625 = 3513302+352+322+362=900+1225+1024+1296=444530^2+35^2+32^2+36^2 = 900+1225+1024+1296 = 4445232+292+282+252=529+841+784+625=277923^2+29^2+28^2+25^2 = 529+841+784+625 = 2779

總平均數 xˉgrand=T1+T2+T3N=117+133+10512=35512≈29.5833\bar{x}_{grand} = \frac{T_1 + T_2 + T_3}{N} = \frac{117 + 133 + 105}{12} = \frac{355}{12} \approx 29.5833

步驟 3:計算變異數分析表 (ANOVA Table)

a. 計算總平方和 (Total Sum of Squares, SST)
SST=∑i=1k∑j=1nixij2−(∑xij)2NSST = \sum_{i=1}^k \sum_{j=1}^{n_i} x_{ij}^2 - \frac{(\sum x_{ij})^2}{N}
∑xij2=3513+4445+2779=10737\sum x_{ij}^2 = 3513 + 4445 + 2779 = 10737
(∑xij)2N=(355)212=12602512≈10502.0833\frac{(\sum x_{ij})^2}{N} = \frac{(355)^2}{12} = \frac{126025}{12} \approx 10502.0833
SST=10737−10502.0833=234.9167SST = 10737 - 10502.0833 = 234.9167

b. 計算組間平方和 (Between Groups Sum of Squares, SSB)
SSB=∑i=1kni(xˉi−xˉgrand)2SSB = \sum_{i=1}^k n_i (\bar{x}_i - \bar{x}_{grand})^2
SSB=4×(29.25−29.5833)2+4×(33.25−29.5833)2+4×(26.25−29.5833)2SSB = 4 \times (29.25 - 29.5833)^2 + 4 \times (33.25 - 29.5833)^2 + 4 \times (26.25 - 29.5833)^2
SSB=4×(−0.3333)2+4×(3.6667)2+4×(−3.3333)2SSB = 4 \times (-0.3333)^2 + 4 \times (3.6667)^2 + 4 \times (-3.3333)^2
SSB=4×0.1111+4×13.4444+4×11.1111SSB = 4 \times 0.1111 + 4 \times 13.4444 + 4 \times 11.1111
SSB=0.4444+53.7776+44.4444=98.6664SSB = 0.4444 + 53.7776 + 44.4444 = 98.6664

另一種計算 SSB 的方法:
SSB=∑i=1kTi2ni−(∑Ti)2NSSB = \sum_{i=1}^k \frac{T_i^2}{n_i} - \frac{(\sum T_i)^2}{N}
SSB=11724+13324+10524−355212SSB = \frac{117^2}{4} + \frac{133^2}{4} + \frac{105^2}{4} - \frac{355^2}{12}
SSB=136894+176894+110254−10502.0833SSB = \frac{13689}{4} + \frac{17689}{4} + \frac{11025}{4} - 10502.0833
SSB=3422.25+4422.25+2756.25−10502.0833SSB = 3422.25 + 4422.25 + 2756.25 - 10502.0833
SSB=10600.75−10502.0833=98.6667SSB = 10600.75 - 10502.0833 = 98.6667 (與前面計算略有差異,因四捨五入)

c. 計算組內平方和 (Within Groups Sum of Squares, SSW)
SSW=SST−SSBSSW = SST - SSB
SSW=234.9167−98.6667=136.25SSW = 234.9167 - 98.6667 = 136.25

🔒

後續完整解題步驟與【答案】

免費註冊,享三天全站完整詳解閱覽。

免費註冊

其他考古題