113 年 國立臺灣海洋大學航運管理學系碩士班乙組《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14 分

  1. A sample of eight observations of variables xx (years of experience) and yy (salary in $1,000s) is shown below:
xx53792468
yy2023151127211714

(a) Calculate and interpret the covariance between xx and yy. (4%)
(b) Calculate the coefficient of correlation, and comment on the relationship between xx and yy. (4%)
(c) Determine the least squares line, and use it to estimate the value of yy for x=6x = 6. (4%)
(d) Draw the scatter diagram and plot the least squares line. (2%)

登入後即可作答並保存紀錄。

這一題的完整詳解

本題主要在考驗對基本敘述統計量,特別是協方差、相關係數、迴歸直線的計算與解釋能力。

首先,我們整理數據並計算所需的統計量:
樣本數 n=8n=8。
∑x=5+3+7+9+2+4+6+8=44\sum x = 5+3+7+9+2+4+6+8 = 44
∑y=20+23+15+11+27+21+17+14=148\sum y = 20+23+15+11+27+21+17+14 = 148
∑x2=52+32+72+92+22+42+62+82=25+9+49+81+4+16+36+64=284\sum x^2 = 5^2+3^2+7^2+9^2+2^2+4^2+6^2+8^2 = 25+9+49+81+4+16+36+64 = 284
∑y2=202+232+152+112+272+212+172+142=400+529+225+121+729+441+289+196=2930\sum y^2 = 20^2+23^2+15^2+11^2+27^2+21^2+17^2+14^2 = 400+529+225+121+729+441+289+196 = 2930
∑xy=(5)(20)+(3)(23)+(7)(15)+(9)(11)+(2)(27)+(4)(21)+(6)(17)+(8)(14)=100+69+105+99+54+84+102+112=725\sum xy = (5)(20)+(3)(23)+(7)(15)+(9)(11)+(2)(27)+(4)(21)+(6)(17)+(8)(14) = 100+69+105+99+54+84+102+112 = 725

計算樣本平均數:
xˉ=∑xn=448=5.5\bar{x} = \frac{\sum x}{n} = \frac{44}{8} = 5.5
yˉ=∑yn=1488=18.5\bar{y} = \frac{\sum y}{n} = \frac{148}{8} = 18.5

計算樣本變異數:
SSx=∑x2−nxˉ2=284−8(5.5)2=284−8(30.25)=284−242=42SS_x = \sum x^2 - n\bar{x}^2 = 284 - 8(5.5)^2 = 284 - 8(30.25) = 284 - 242 = 42
sx2=SSxn−1=427=6s_x^2 = \frac{SS_x}{n-1} = \frac{42}{7} = 6

SSy=∑y2−nyˉ2=2930−8(18.5)2=2930−8(342.25)=2930−2738=192SS_y = \sum y^2 - n\bar{y}^2 = 2930 - 8(18.5)^2 = 2930 - 8(342.25) = 2930 - 2738 = 192
sy2=SSyn−1=1927≈27.4286s_y^2 = \frac{SS_y}{n-1} = \frac{192}{7} \approx 27.4286

(a) Calculate and interpret the covariance between xx and yy. (4%)
計算樣本共變異數 sxys_{xy}:
SSxy=∑xy−nxˉyˉ=725−8(5.5)(18.5)=725−8(101.75)=725−814=−89SS_{xy} = \sum xy - n\bar{x}\bar{y} = 725 - 8(5.5)(18.5) = 725 - 8(101.75) = 725 - 814 = -89
sxy=SSxyn−1=−897≈−12.714s_{xy} = \frac{SS_{xy}}{n-1} = \frac{-89}{7} \approx -12.714

解釋:
樣本共變異數為 -12.714。由於共變異數為負值,表示變數 xx (年資) 與 yy (薪資) 之間存在負向關係,即年資越長,薪資傾向越低,這與一般認知相反,可能表示樣本中存在其他影響因素,或是該樣本的特殊性。

(b) Calculate the coefficient of correlation, and comment on the relationship between xx and yy. (4%)
計算樣本相關係數 rr:

🔒

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12 分

  1. Three messenger services deliver to a small town in Oregon. Service A has 60% of all the scheduled deliveries, service B has 30%, and service C has the remaining 10%. Their on-time rates are 80%, 60%, and 40% respectively. Define event OO as a service delivers a package on time. (12%)

(a) Calculate P(A and O)P(A \text{ and } O) (4%)
(b) If a package was delivered on time, what is the probability that it was service B? (4%)
(c) If a package was delivered 40 minutes late, what is the probability that it was service C? (4%)

登入後即可作答並保存紀錄。

這一題的完整詳解

此題為機率與條件機率的應用題,涉及聯合機率、貝氏定理。

首先,我們定義事件與機率:
AA: 該次配送由服務 A 負責,P(A)=0.60P(A) = 0.60
BB: 該次配送由服務 B 負責,P(B)=0.30P(B) = 0.30
CC: 該次配送由服務 C 負責,P(C)=0.10P(C) = 0.10
OO: 該次配送準時完成。

已知各服務的準時率(條件機率):
P(O∣A)=0.80P(O|A) = 0.80 (服務 A 準時的機率)
P(O∣B)=0.60P(O|B) = 0.60 (服務 B 準時的機率)
P(O∣C)=0.40P(O|C) = 0.40 (服務 C 準時的機率)

(a) Calculate P(A and O)P(A \text{ and } O) (4%)
計算服務 A 負責且準時送達的聯合機率。
根據聯合機率的定義:
P(A and O)=P(A∩O)=P(O∣A)P(A)P(A \text{ and } O) = P(A \cap O) = P(O|A) P(A)
P(A∩O)=(0.80)(0.60)=0.48P(A \cap O) = (0.80)(0.60) = 0.48

【答案】P(A and O)=0.48P(A \text{ and } O) = 0.48

(b) If a package was delivered on time, what is the probability that it was service B? (4%)
這是求條件機率 P(B∣O)P(B|O)。
根據條件機率的定義:
P(B∣O)=P(B and O)P(O)P(B|O) = \frac{P(B \text{ and } O)}{P(O)}

首先,我們需要計算 P(O)P(O),即任意一次配送準時送達的總機率。根據全機率法則:
P(O)=P(O∣A)P(A)+P(O∣B)P(B)+P(O∣C)P(C)P(O) = P(O|A)P(A) + P(O|B)P(B) + P(O|C)P(C)
P(O)=(0.80)(0.60)+(0.60)(0.30)+(0.40)(0.10)P(O) = (0.80)(0.60) + (0.60)(0.30) + (0.40)(0.10)
P(O)=0.48+0.18+0.04=0.70P(O) = 0.48 + 0.18 + 0.04 = 0.70

現在可以計算 P(B∣O)P(B|O):
P(B and O)=P(O∣B)P(B)=(0.60)(0.30)=0.18P(B \text{ and } O) = P(O|B)P(B) = (0.60)(0.30) = 0.18

🔒

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8 分

  1. The number of ship arrivals at a berth between 3:00 and 5:00 P.M. has a Poisson distribution with a mean of 12. (8%)

(a) Find the probability that the number of arrivals between 3:30 and 4:00 P.M. is at least 10. (4%)
(b) Find the probability that the number of arrivals between 4:00 and 5:00 P.M. is exactly two. (4%)

登入後即可作答並保存紀錄。

這一題的完整詳解

本題考驗對 Poisson 分布的理解與應用,特別是時間區間的轉換與機率計算。

題目說明:兩小時 (3:00 P.M. 到 5:00 P.M.) 的船隻到港數服從 Poisson 分布,平均數為 12。
設 N(t1,t2)N(t_1, t_2) 表示在時間區間 (t1,t2)(t_1, t_2) 內的到港數。
已知在 2 小時內,平均到港數為 λ2hr=12\lambda_{2hr} = 12。

Poisson 分布的特性是事件發生的平均率是恆定的。因此,我們可以推算出不同時間區間的平均到港數。
平均到港率 λrate=12 arrivals2 hours=6 arrivals per hour\lambda_{rate} = \frac{12 \text{ arrivals}}{2 \text{ hours}} = 6 \text{ arrivals per hour}。

(a) Find the probability that the number of arrivals between 3:30 and 4:00 P.M. is at least 10. (4%)
時間區間是 3:30 P.M. 到 4:00 P.M.,長度為 0.5 小時。
在此 0.5 小時區間內的平均到港數為:
λ0.5hr=λrate×0.5 hours=6×0.5=3\lambda_{0.5hr} = \lambda_{rate} \times 0.5 \text{ hours} = 6 \times 0.5 = 3。
所以,在這個 0.5 小時區間內,到港數 X∼Poisson(λ=3)X \sim \text{Poisson}(\lambda=3)。

我們要計算 P(X≥10)P(X \ge 10)。
P(X≥10)=1−P(X≤9)P(X \ge 10) = 1 - P(X \le 9)
P(X≤9)=∑k=09e−33kk!P(X \le 9) = \sum_{k=0}^{9} \frac{e^{-3} 3^k}{k!}

直接計算會比較複雜,通常需要查表或使用統計軟體。
P(X≤9)=P(X=0)+P(X=1)+...+P(X=9)P(X \le 9) = P(X=0) + P(X=1) + ... + P(X=9)
P(X=0)=e−3300!≈0.0498P(X=0) = e^{-3} \frac{3^0}{0!} \approx 0.0498
P(X=1)=e−3311!≈0.1494P(X=1) = e^{-3} \frac{3^1}{1!} \approx 0.1494
P(X=2)=e−3322!≈0.2240P(X=2) = e^{-3} \frac{3^2}{2!} \approx 0.2240
P(X=3)=e−3333!≈0.2240P(X=3) = e^{-3} \frac{3^3}{3!} \approx 0.2240

🔒

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8 分

  1. The recent average starting salary for new college graduates in IT systems is 47,500.Assumesalariesarenormallydistributedwithastandarddeviationof47,500. Assume salaries are normally distributed with a standard deviation of 4,500. (8%)

(a) What percent of starting salaries are no more than $42,250? (4%)
(b) What is the cutoff for the top 3% of the salaries? (4%)

登入後即可作答並保存紀錄。

這一題的完整詳解

核心觀念

本題評量連續隨機變數之常態分配(Normal Distribution)及其標準化(Standardization)應用,涵蓋以下關鍵觀念與公式:

  1. 常態分配之定義與符號:
    設隨機變數 XX 代表 IT 系統新進大學畢業生的起薪,則:

    X∼N(μ,σ2)X \sim N(\mu, \sigma^2)

    其中母體平均數 μ=47500\mu = 47500,母體標準差 σ=4500\sigma = 4500。

  2. 標準化轉換(ZZ-Transformation):
    欲求常態隨機變數之機率或分位數,需透過標準化公式將其轉換為標準常態隨機變數 Z∼N(0,1)Z \sim N(0, 1):

    Z=X−μσZ = \frac{X - \mu}{\sigma}
  3. 百分位數與臨界值(Cutoff / Quantile)之逆轉換:
    若已知機率 P(X≥x0)=αP(X \ge x_0) = \alpha(即累積機率 P(X≤x0)=1−αP(X \le x_0) = 1 - \alpha),由標準常態分配表查得對應的臨界值 zαz_\alpha 後,可藉由逆運算求得原變數之門檻值:

    x0=μ+zα⋅σx_0 = \mu + z_\alpha \cdot \sigma

解題方法

(a) 計算起薪不超過 $42,250 的百分比

步驟一:設定事件機率並進行標準化
欲求 P(X≤42250)P(X \le 42250):

Z=42250−475004500=−52504500=−76≈−1.17Z = \frac{42250 - 47500}{4500} = \frac{-5250}{4500} = -\frac{7}{6} \approx -1.17

步驟二:查標準常態分配累積機率表(Φ(z)\Phi(z))

P(X≤42250)=P(Z≤−1.17)=1−Φ(1.17)=1−0.8790=0.1210P(X \le 42250) = P(Z \le -1.17) = 1 - \Phi(1.17) = 1 - 0.8790 = 0.1210

(若以精確值 Z=−1.1667Z = -1.1667 線性內插計算,Φ(−1.1667)≈0.1217\Phi(-1.1667) \approx 0.1217 即 12.17%12.17\%)

🔒

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8 分

  1. A survey of banks estimated the following probabilities for the interest rate being charged on a equity loan based on a 30-year loan, based on past records. (8%)
Interest Rate6.0%6.5%7.0%7.5%>7.5%
Probability0.200.230.250.28.04

a. If a bank is selected at random from this distribution, what is the probability that the interest rate charged on a home loan exceeds 7.0%? (4%)
b. What is the most common interest rate? (4%)

登入後即可作答並保存紀錄。

這一題的完整詳解

本題考驗對離散機率分布的理解,包括機率計算和眾數的判斷。

題目給出了一個離散機率分布,描述了銀行對 30 年期股權貸款 (equity loan) 的利率。
我們假設這個分布也適用於房貸 (home loan),或者題目意在考驗對給定機率分布的讀取能力。

(a) If a bank is selected at random from this distribution, what is the probability that the interest rate charged on a home loan exceeds 7.0%? (4%)
我們要找的機率是 P(Interest Rate>7.0%)P(\text{Interest Rate} > 7.0\%)。
從表格中,大於 7.0% 的利率類別是 7.5% 和 >7.5%。
因此,我們需要將這兩個類別的機率相加:

🔒

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 一 題12 分

第二部份
(求解過程中,凡遇 +、-、×、÷、√一律取小數第三位,第四位四捨五入)

  1. A survey sampled men and women workers and asked if they expected to get a raise or promotion this year. Suppose the survey sampled 200 men and 200 women. If 104 of the men replied Yes and 74 of the women replied Yes, are the results statistically significant in that you can conclude a greater proportion of men are expecting to get a raise or a promotion this year?
    (a) (4%) State the hypothesis test in terms of the population proportion of men and the population proportion of women?
    (b) (8%) Use a 0.01 level of significance. What is the p-value and what is your conclusion?

登入後即可作答並保存紀錄。

這一題的完整詳解

本題考驗對比例的假設檢定 (Hypothesis Test for Proportions),特別是獨立樣本比例的比較。

題目情境:
樣本大小:男人 nm=200n_m = 200,女人 nw=200n_w = 200。
成功次數:男人回答 Yes xm=104x_m = 104,女人回答 Yes xw=74x_w = 74。
顯著水準 α=0.01\alpha = 0.01。
檢定目的:判斷男性預期加薪/升遷的比例是否顯著大於女性。

(a) State the hypothesis test in terms of the population proportion of men and the population proportion of women? (4%)
我們需要建立虛無假設 (H0H_0) 和對立假設 (H1H_1)。
令 pmp_m 為男性預期加薪/升遷的總體比例。
令 pwp_w 為女性預期加薪/升遷的總體比例。

虛無假設 (H0H_0):男性與女性的比例沒有差異,或者男性的比例不顯著大於女性。
H0:pm≤pwH_0: p_m \le p_w 或者 H0:pm−pw≤0H_0: p_m - p_w \le 0

對立假設 (H1H_1):男性的比例顯著大於女性。
H1:pm>pwH_1: p_m > p_w 或者 H1:pm−pw>0H_1: p_m - p_w > 0

這是一個單尾檢定 (right-tailed test)。

【答案】
H0:pm≤pwH_0: p_m \le p_w
H1:pm>pwH_1: p_m > p_w

(b) Use a 0.01 level of significance. What is the p-value and what is your conclusion? (8%)
首先,計算樣本比例:
p^m=xmnm=104200=0.52\hat{p}_m = \frac{x_m}{n_m} = \frac{104}{200} = 0.52
p^w=xwnw=74200=0.37\hat{p}_w = \frac{x_w}{n_w} = \frac{74}{200} = 0.37

在虛無假設 H0:pm=pwH_0: p_m = p_w 下,我們需要計算一個合併的樣本比例 p^pooled\hat{p}_{pooled} 來計算標準誤。

🔒

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 二 題21 分

  1. An important factor in selecting software for word-processing and database management systems is the time (in hours) required to learn how to use the system. To evaluate five file management systems, a firm designed a test involving three word-processing operators. Each of the three operators was trained on each of the five file management systems. The data obtained below.
    (a) (11%) Construct an analysis of variance table.
    (b) (5%) Use α=0.05\alpha=0.05 to test for any difference in the mean training time for the five systems.
    (c) (5%) Use α=0.05\alpha=0.05 to test for any difference in the mean training time for the three operators.
systemoperatorABCDE
16972596575
27073647168
37168636876

登入後即可作答並保存紀錄。

這一題的完整詳解

核心觀念

本題比較五種檔案管理系統的平均學習時間,但三位操作員的學習速度可能不同,因此採用「隨機完全區集設計」的二因子變異數分析:

  • 處理因子:檔案管理系統,共 t=5t=5 種。
  • 區集因子:操作員,共 b=3b=3 位。
  • 每位操作員均使用五種系統,因此每個系統皆有三筆觀測值。
  • 假設模型為
Yij=μ+τi+βj+εij,Y_{ij}=\mu+\tau_i+\beta_j+\varepsilon_{ij},

其中 τi\tau_i 表示系統效果,βj\beta_j 表示操作員效果。

本題未提供重複觀測,因此不另行估計系統與操作員的交互作用,誤差項包含隨機誤差及未估計的交互作用。


解題方法

將題目表格解讀為:欄位 AA 至 EE 代表五種系統,列 11 至 33 代表三位操作員。

1. 基本總計

各操作員總和為

R1=69+72+59+65+75=340,R_1=69+72+59+65+75=340, R2=70+73+64+71+68=346,R_2=70+73+64+71+68=346, R3=71+68+63+68+76=346.R_3=71+68+63+68+76=346.

各系統總和為

CA=210,CB=213,CC=186,CD=204,CE=219.\begin{aligned} C_A&=210,\\ C_B&=213,\\ C_C&=186,\\ C_D&=204,\\ C_E&=219. \end{aligned}

總和為

T=1032,T=1032,

觀測值總數為

N=3×5=15.N=3\times5=15.

校正項為

CT=T2N=1032215=71001.6.CT=\frac{T^2}{N} =\frac{1032^2}{15} =71001.6.

2. 總平方和

全部觀測值平方和為

∑yij2=71300.\sum y_{ij}^2=71300.

因此

SST=∑yij2−CT=71300−71001.6=298.4.SS_T=\sum y_{ij}^2-CT =71300-71001.6 =298.4.

3. 系統平方和

五種系統各有 b=3b=3 筆資料,因此

SS系統=2102+2132+1862+2042+21923−CT.SS_{\text{系統}} =\frac{210^2+213^2+186^2+204^2+219^2}{3}-CT. SS系統=2136423−71001.6=212.4.SS_{\text{系統}} =\frac{213642}{3}-71001.6 =212.4.

4. 操作員平方和

三位操作員各有 t=5t=5 筆資料,因此

SS操作員=3402+3462+34625−CT.SS_{\text{操作員}} =\frac{340^2+346^2+346^2}{5}-CT. SS操作員=3550325−71001.6=4.8.SS_{\text{操作員}} =\frac{355032}{5}-71001.6 =4.8.

5. 誤差平方和

SSE=SST−SS系統−SS操作員SS_E =SS_T-SS_{\text{系統}}-SS_{\text{操作員}} SSE=298.4−212.4−4.8=81.2.SS_E=298.4-212.4-4.8=81.2.

(a) 變異數分析表

自由度如下:

  • 總和:N−1=15−1=14N-1=15-1=14
  • 系統:t−1=5−1=4t-1=5-1=4
  • 操作員:b−1=3−1=2b-1=3-1=2
  • 誤差:(t−1)(b−1)=4×2=8(t-1)(b-1)=4\times2=8

均方為

MS系統=212.44=53.10,MS_{\text{系統}}=\frac{212.4}{4}=53.10, MS操作員=4.82=2.40,MS_{\text{操作員}}=\frac{4.8}{2}=2.40, MSE=81.28=10.15.MS_E=\frac{81.2}{8}=10.15.

因此檢定統計量為

F系統=MS系統MSE=53.1010.15=5.23,F_{\text{系統}} =\frac{MS_{\text{系統}}}{MS_E} =\frac{53.10}{10.15} =5.23, F操作員=MS操作員MSE=2.4010.15=0.24.F_{\text{操作員}} =\frac{MS_{\text{操作員}}}{MS_E} =\frac{2.40}{10.15} =0.24.
變異來源平方和 SSSS自由度 dfdf均方 MSMSFF 值
系統212.4212.44453.1053.105.235.23
操作員4.84.8222.402.400.240.24
🔒

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17 分

  1. A training center, wishing to demonstrate the effectiveness of their methods, tests 12 of their clients after different numbers of days(x) of training, recording their scores(y) on the sample test. statistical information as below,
    x=15.4x=15.4, yˉ=89.2\bar{y} = 89.2, SSX=142.5SSX =142.5, SSY=650.3SSY =650.3, SSXY=268.7SSXY =268.7

(a) (4%) Find the best-fitting line for the above data.
(b) (8%) Do the data provide sufficient evidence to indicate that a client would get a higher score if the training time is longer? Test using α=0.05\alpha=0.05.
(c) (5%) Develop a 95% confidence interval for the mean score with a 14-day-long training time.

登入後即可作答並保存紀錄。

這一題的完整詳解

本題考驗對簡單線性迴歸的應用,包括迴歸直線的建立、迴歸係數的假設檢定,以及迴歸線的信賴區間估計。

已知數據:
樣本數 n=12n=12。
xx 的樣本平均數 xˉ=15.4\bar{x} = 15.4 (題目直接給出,通常應由 ∑x\sum x 和 nn 計算)
yy 的樣本平均數 yˉ=89.2\bar{y} = 89.2
xx 的平方和 SSX=142.5SSX = 142.5
yy 的平方和 SSY=650.3SSY = 650.3
xyxy 的交叉乘積和 SSXY=268.7SSXY = 268.7

(a) Find the best-fitting line for the above data. (4%)
最佳擬合線 (迴歸直線) 的形式為 y^=b0+b1x\hat{y} = b_0 + b_1 x。
首先計算斜率 b1b_1:
b1=SSXYSSX=268.7142.5≈1.8856b_1 = \frac{SSXY}{SSX} = \frac{268.7}{142.5} \approx 1.8856

接著計算截距 b0b_0:
b0=yˉ−b1xˉb_0 = \bar{y} - b_1 \bar{x}
b0=89.2−(1.8856)(15.4)b_0 = 89.2 - (1.8856)(15.4)
b0=89.2−29.03864≈60.16136b_0 = 89.2 - 29.03864 \approx 60.16136

迴歸直線方程式為 y^≈60.161+1.886x\hat{y} \approx 60.161 + 1.886 x。

【答案】最佳擬合線為 y^=60.161+1.886x\hat{y} = 60.161 + 1.886 x。

(b) Do the data provide sufficient evidence to indicate that a client would get a higher score if the training time is longer? Test using α=0.05\alpha=0.05. (8%)
這是檢定迴歸直線的斜率 b1b_1 是否顯著大於零。
虛無假設 H0:β1=0H_0: \beta_1 = 0 (訓練時間與得分之間沒有線性關係,或訓練時間越長得分不越高)
對立假設 H1:β1>0H_1: \beta_1 > 0 (訓練時間越長,得分越高)
這是一個單尾檢定。

首先,我們需要計算檢定統計量 tt。
t=b1−β1,0sb1t = \frac{b_1 - \beta_{1,0}}{s_{b_1}}
其中 β1,0=0\beta_{1,0} = 0 (來自 H0H_0), sb1s_{b_1} 是斜率 b1b_1 的標準誤。
sb1=seSSXs_{b_1} = \frac{s_e}{\sqrt{SSX}}
其中 ses_e 是誤差標準差。
se2=SSEn−2s_e^2 = \frac{SSE}{n-2}
SSE=SSY−b1SSXY=650.3−(1.8856)(268.7)=650.3−506.57≈143.73SSE = SSY - b_1 SSXY = 650.3 - (1.8856)(268.7) = 650.3 - 506.57 \approx 143.73
se2=143.7312−2=143.7310=14.373s_e^2 = \frac{143.73}{12-2} = \frac{143.73}{10} = 14.373
se=14.373≈3.791s_e = \sqrt{14.373} \approx 3.791

sb1=3.791142.5=3.79111.937≈0.3176s_{b_1} = \frac{3.791}{\sqrt{142.5}} = \frac{3.791}{11.937} \approx 0.3176

計算 tt 統計量:
t=1.8856−00.3176≈5.937t = \frac{1.8856 - 0}{0.3176} \approx 5.937

自由度 df=n−2=12−2=10df = n-2 = 12-2 = 10。
顯著水準 α=0.05\alpha = 0.05。

🔒

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8 分

  1. The recent average starting salary for new college graduates in IT systems is 47,500.Assumesalariesarenormallydistributedwithastandarddeviationof47,500. Assume salaries are normally distributed with a standard deviation of 4,500. (8%)

(a) What percent of starting salaries are no more than $42,250? (4%)
(b) What is the cutoff for the top 3% of the salaries? (4%)

登入後即可作答並保存紀錄。

這一題的完整詳解

核心觀念

本題評量連續隨機變數之常態分配(Normal Distribution)及其標準化(Standardization)應用,涵蓋以下關鍵觀念與公式:

  1. 常態分配之定義與符號:
    設隨機變數 XX 代表 IT 系統新進大學畢業生的起薪,則:

    X∼N(μ,σ2)X \sim N(\mu, \sigma^2)

    其中母體平均數 μ=47500\mu = 47500,母體標準差 σ=4500\sigma = 4500。

  2. 標準化轉換(ZZ-Transformation):
    欲求常態隨機變數之機率或分位數,需透過標準化公式將其轉換為標準常態隨機變數 Z∼N(0,1)Z \sim N(0, 1):

    Z=X−μσZ = \frac{X - \mu}{\sigma}
  3. 百分位數與臨界值(Cutoff / Quantile)之逆轉換:
    若已知機率 P(X≥x0)=αP(X \ge x_0) = \alpha(即累積機率 P(X≤x0)=1−αP(X \le x_0) = 1 - \alpha),由標準常態分配表查得對應的臨界值 zαz_\alpha 後,可藉由逆運算求得原變數之門檻值:

    x0=μ+zα⋅σx_0 = \mu + z_\alpha \cdot \sigma

解題方法

(a) 計算起薪不超過 $42,250 的百分比

步驟一:設定事件機率並進行標準化
欲求 P(X≤42250)P(X \le 42250):

Z=42250−475004500=−52504500=−76≈−1.17Z = \frac{42250 - 47500}{4500} = \frac{-5250}{4500} = -\frac{7}{6} \approx -1.17

步驟二:查標準常態分配累積機率表(Φ(z)\Phi(z))

P(X≤42250)=P(Z≤−1.17)=1−Φ(1.17)=1−0.8790=0.1210P(X \le 42250) = P(Z \le -1.17) = 1 - \Phi(1.17) = 1 - 0.8790 = 0.1210

(若以精確值 Z=−1.1667Z = -1.1667 線性內插計算,Φ(−1.1667)≈0.1217\Phi(-1.1667) \approx 0.1217 即 12.17%12.17\%)

🔒

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8 分

  1. A survey of banks estimated the following probabilities for the interest rate being charged on a equity loan based on a 30-year loan, based on past records. (8%)
Interest Rate6.0%6.5%7.0%7.5%>7.5%
Probability0.200.230.250.28.04

a. If a bank is selected at random from this distribution, what is the probability that the interest rate charged on a home loan exceeds 7.0%? (4%)
b. What is the most common interest rate? (4%)

登入後即可作答並保存紀錄。

這一題的完整詳解

本題考驗對離散機率分布的理解,包括機率計算和眾數的判斷。

題目給出了一個離散機率分布,描述了銀行對 30 年期股權貸款 (equity loan) 的利率。
我們假設這個分布也適用於房貸 (home loan),或者題目意在考驗對給定機率分布的讀取能力。

(a) If a bank is selected at random from this distribution, what is the probability that the interest rate charged on a home loan exceeds 7.0%? (4%)
我們要找的機率是 P(Interest Rate>7.0%)P(\text{Interest Rate} > 7.0\%)。
從表格中,大於 7.0% 的利率類別是 7.5% 和 >7.5%。
因此,我們需要將這兩個類別的機率相加:

🔒

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