112 年 國立中山大學企業管理學系碩士甲班乙組《商用統計學》

📄 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

Use the tables of probability distributions in Appendix as needed (page 6-9).
Which of the following descriptions is/are correct?
(A) "Age group" is a categorical variable;
(B) "Temperature in Celsius" is a ratio variable;
(C) Responding to "How likely do you recommend EVA Airline to others?" after a flight contains the following options in a questionnaire "Definitely Will, Probably Will, May or May Not, Probably Will Not, Definitely Will Not." This is a numerical data;
(D) "Descriptive Statistics" is the process of using data obtained from a sample to make estimates or test hypotheses about the characteristics of a population.;
(E) "Data Collection" does not belong to the field of statistics.

登入後即可作答並保存紀錄。

這一題的完整詳解

此題主要在考察統計學中關於資料類型、敘述統計、資料蒐集等基本概念的理解。

(A) 年齡層 (Age group) 屬於類別變數 (categorical variable),因為它是將受試者分成不同的群組,例如年輕、中年、老年等,這些群組沒有數值上的順序或大小之分,即使我們可能會用數字代表,但其本質是類別。因此,此敘述正確。

(B) 以攝氏 (Celsius) 為單位的溫度是一個比例變數 (ratio variable)。比例變數具有固定的零點,且數值之間的比例有意義。例如,0°C 表示沒有溫度(但這與絕對零度不同,絕對零度才是比例變數的固定零點),而 20°C 是 10°C 的兩倍,這在比例變數下是有意義的。然而,攝氏溫度的零點並非絕對零度,因此嚴格來說,攝氏溫度是區間變數 (interval variable),而不是比例變數 (ratio variable)。如果題目意指絕對溫度 (Kelvin),那會是比例變數。

🔒

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

Which of the following is/are correct?
(A) Stem-and-Leaf plot can be used to summarize one categorical variable with multiple categories;
(B) Scatter diagram is used to visualize the potential association between two numerical variables;
(C) A "pie chart" is good for comparing the means of a numerical variable between two independent groups;
(D) To visualize a distribution for a numerical variable, we can use "histogram";
(E) In a "Box-plot", the line drawn in the box indicates the location of the mean.

登入後即可作答並保存紀錄。

這一題的完整詳解

此題主要考察各種統計圖表的用途與特性。

(A) Stem-and-Leaf plot(莖葉圖)主要用於展示數值資料的分佈情況,特別是對於單一數值變數。它能同時顯示資料的集中趨勢、離散趨勢以及個別資料點。莖葉圖不適用於類別變數 (categorical variable)。因此,此敘述錯誤。

(B) Scatter diagram(散佈圖)是用來觀察兩個數值變數 (numerical variables) 之間關係的一種圖形。它能幫助我們判斷兩個變數是否存在線性關聯、正相關、負相關或無關聯,並可大致估計關聯的強度。因此,此敘述正確。

🔒

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

Which of the following is/are used to measure variability?
(A) Geometric mean;
(B) Interquartile range;
(C) Coefficient of variation;
(D) Standard deviation;
(E) Midrange.

登入後即可作答並保存紀錄。

這一題的完整詳解

此題在考察用來衡量資料離散程度(變異性)的統計量。

(A) Geometric mean(幾何平均數)是一種用來計算一組數值平均值的指標,特別適用於計算成長率、利率等連乘的平均。它與變異性無關。因此,此敘述錯誤。

(B) Interquartile range(四分位距,IQR)是指第三四分位數 (Q3) 與第一四分位數 (Q1) 之差 (Q3−Q1Q_3 - Q_1)。它衡量了中間 50% 資料的離散程度,是衡量變異性的指標之一,尤其對離群值較不敏感。因此,此敘述正確。

(C) Coefficient of variation(變異係數,CV)是以百分比表示的相對變異量,計算方式為標準差除以平均數 (sxˉ\frac{s}{\bar{x}} 或

🔒

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

Assume P(X) is the probability of event X and P(Y) is the probability of event Y. Which of the following descriptions is/are correct?
(A) P(X∩Y)=P(X).P(Y) if X and Y are mutually exclusive;
(B) P(Y|X)=P(X) if X and Y are independent events;
(C) If X and Y are mutually exclusive, Y is the complement of X;
(D) If X and Y are independent, P(XUY)=P(X)·P(Y);
(E) None of above.

登入後即可作答並保存紀錄。

這一題的完整詳解

此題考察機率論中的基本概念,包括獨立事件、互斥事件、條件機率、聯合機率與聯集機率的關係。

(A) 如果 X 和 Y 是互斥事件 (mutually exclusive),則表示它們不能同時發生,即 P(X∩Y)=0P(X \cap Y) = 0。而 P(X)⋅P(Y)P(X) \cdot P(Y) 通常只有在 X 和 Y 是獨立事件 (independent) 且 P(X)>0,P(Y)>0P(X)>0, P(Y)>0 時才等於 P(X∩Y)P(X \cap Y)。因此,互斥事件不代表 P(X∩Y)=P(X)⋅P(Y)P(X \cap Y) = P(X) \cdot P(Y)。此敘述錯誤。

(B) 如果 X 和 Y 是獨立事件 (independent),則表示一個事件的發生不會影響另一個事件的發生機率。條件機率的定義是 P(Y∣X)=P(X∩Y)P(X)P(Y|X) = \frac{P(X \cap Y)}{P(X)}。對於獨立事件,我們知道 P(X∩Y)=P(X)⋅P(Y)P(X \cap Y) = P(X) \cdot P(Y)。因此,P(Y∣X)=P(X)⋅P(Y)P(X)=P(Y)P(Y|X) = \frac{P(X) \cdot P(Y)}{P(X)} = P(Y)。所以,若 X 和 Y 獨立,則 P(Y∣X)=P(Y)P(Y|X) = P(Y),而不是 P(X)P(X)。除非 X 和 Y 的機率恰好相等,否則此敘述錯誤。

🔒

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 5 題

A personal care company ran two television advertisements for one of its new shampoo product.
On the basis of a survey that was conducted, probabilities were assigned to the following events:
M= individual purchased the product
S₁= individual recalls seeing the advertisement 1
S₂= individual recalls seeing the advertisement 2
MS₁ = individual purchased the product and recalls seeing the advertisement 1
MS₂ = individual purchased the product and recalls seeing the advertisement 2
The probabilities assigned were P(M) = 0.2, P(S₁) = 0.4, P(S₂) = 0.3, P(MS₁) = 0.12, and P(MS₂) = 0.1. Which of the following is/are correct?
(A) The probability of an individual's purchasing the product given that the individual recalls seeing the advertisement 1 is 0.3;
(B) The probability of an individual's purchasing the product given that the individual recalls seeing the advertisement 2 is 0.1;
(C) Seeing the advertisement 2 does not increase the probability that the individual will purchase the product;
(D) Assume that individuals who do not purchase the company's shampoo product buy from its competitors. Based on the calculation, continuing the advertisement 1 should increase the market share;
(E) The advertisement 2 has a bigger effect on consumer purchases.

登入後即可作答並保存紀錄。

這一題的完整詳解

此題考察條件機率、獨立性檢定以及對行銷活動效果的初步評估。

首先,我們將已知的機率列出:
P(M)=0.2P(M) = 0.2 (購買產品的機率)
P(S1)=0.4P(S_1) = 0.4 (看到廣告 1 的機率)
P(S2)=0.3P(S_2) = 0.3 (看到廣告 2 的機率)
P(MS1)=0.12P(MS_1) = 0.12 (購買產品且看到廣告 1 的機率)
P(MS2)=0.1P(MS_2) = 0.1 (購買產品且看到廣告 2 的機率)

(A) 計算「在看過廣告 1 的條件下,購買產品的機率」。這是一個條件機率:
P(M∣S1)=P(MS1)P(S1)=0.120.4=0.3P(M|S_1) = \frac{P(MS_1)}{P(S_1)} = \frac{0.12}{0.4} = 0.3
此敘述正確。

(B) 計算「在看過廣告 2 的條件下,購買產品的機率」。這也是一個條件機率:
P(M∣S2)=P(MS2)P(S2)=0.10.3=13≈0.3333P(M|S_2) = \frac{P(MS_2)}{P(S_2)} = \frac{0.1}{0.3} = \frac{1}{3} \approx 0.3333
題目中說該機率為 0.1,這是錯誤的。因此,此敘述錯誤。

(C) 判斷「看到廣告 2 是否增加購買產品的機率」。我們可以比較 P(M∣S2)P(M|S_2) 與 P(M)P(M)。
P(M∣S2)=1/3≈0.3333P(M|S_2) = 1/3 \approx 0.3333

🔒

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 6 題

A consulting firm submitted a bid for a large research project. The firm's management initially
felt they had a 50% chance of getting the project. However, the agency to which the bid was
submitted subsequently requested additional information on the bid. Past experience indicates that
for 75% of the successful bids and 40% of the unsuccessful bids the agency requested additional
information. Which of the following is/are correct?
(A) The prior probability of the bid being successful is 0.5;
(B) The conditional probability of a request for additional information given that the bid will ultimately be successful is 0.25;
(C) The posterior probability that the bid will be successful given a request for additional information is 0.65;
(D) All above are correct;
(E) None of the above is correct.

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:

  • 先驗機率(prior probability)
  • 條件機率(conditional probability)
  • 貝氏定理(Bayes’ theorem)
  • 聯合機率與全機率公式

設:

  • SS:投標成功
  • ScS^c:投標失敗
  • RR:機關要求補充投標資訊

題目給定:

P(S)=0.5,P(Sc)=0.5P(S)=0.5,\qquad P(S^c)=0.5 P(R∣S)=0.75,P(R∣Sc)=0.40P(R\mid S)=0.75,\qquad P(R\mid S^c)=0.40

解題方法

題目已直接給出管理階層原先對成功機率的判斷,因此先驗機率為:

P(S)=0.5P(S)=0.5

當機關要求補充資訊後,要計算投標成功的後驗機率 P(S∣R)P(S\mid R),使用貝氏定理:

P(S∣R)=P(R∣S)P(S)P(R)P(S\mid R) = \frac{P(R\mid S)P(S)} {P(R)}

其中 P(R)P(R) 可由全機率公式求得:

P(R)=P(R∣S)P(S)+P(R∣Sc)P(Sc)P(R)=P(R\mid S)P(S)+P(R\mid S^c)P(S^c)

代入數值:

P(R)=(0.75)(0.5)+(0.40)(0.5)=0.375+0.20=0.575P(R) = (0.75)(0.5)+(0.40)(0.5) =0.375+0.20 =0.575

因此:

P(S∣R)=(0.75)(0.5)0.575=0.3750.575≈0.6522P(S\mid R) = \frac{(0.75)(0.5)}{0.575} = \frac{0.375}{0.575} \approx 0.6522

故要求補充資訊後,投標成功的後驗機率約為 0.650.65。


選項分析

(A)The prior probability of the bid being successful is 0.5

先驗機率是尚未考慮新資訊前,投標成功的機率。

題目明確指出管理階層原先認為成功機率為 50%50\%,因此:

P(S)=0.5P(S)=0.5

此選項正確。


(B)The conditional probability of a request for additional information given that the bid will ultimately be successful is 0.25

此選項所描述的是:

🔒

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 7 題

Which of the following belongs to continuous probability distributions?
(A) Normal probability distribution;
(B) Poisson probability distribution;
(C) Binomial probability distribution;
(D) Exponential probability distribution;
(E) Chi-square probability distribution.

登入後即可作答並保存紀錄。

這一題的完整詳解

此題在考察常見機率分佈的分類,區分連續型 (continuous) 和離散型 (discrete) 機率分佈。

連續型機率分佈是指其機率密度函數 (probability density function, PDF) 定義在一個連續的範圍上,變數可以取區間內的任何值。
離散型機率分佈是指其機率質量函數 (probability mass function, PMF) 定義在離散點上,變數只能取有限個或可數無限個值。

我們逐一檢視各選項:
(A) Normal probability distribution (常態分佈):這是一個典型的連續型機率分佈,其機率密度函數在整個實數線上連續。因此,(A) 正確。

(B) Poisson probability distribution (卜瓦松分佈):這是一個離散型機率分佈,用於描述在固定時間或空間間隔內,某事件發生次數的機率。事件發生的次數是整數(0, 1, 2, ...),是離散的。

🔒

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 8 題

Customers arrive at a bank randomly and independently. The probability of an arrival is the same for any interval of equal length. The mean time interval of arrival between two customers is 100 seconds. Which of the following is/are correct?
(A) The probability that less than 2 customers will arrive during any given 5 minutes of operation is 0.0199;
(B) The probability that 4 or more customers will arrive during any given 5 minutes of operation is 0.3528;
(C) The probability that more than 6 customers will arrive during any given 10 minutes of operation is 0.7440;
(D) The probability that exact 3 customers will arrive during any given 15 minutes of operation is 0.0212;
(E) The standard deviation of the arrival rate per 10 minutes is 6.

登入後即可作答並保存紀錄。

這一題的完整詳解

核心觀念

顧客以隨機、彼此獨立的方式到達,且等長時間區間的到達機率相同,符合卜瓦松過程的假設。因此,固定時間內的顧客到達數 XX 服從卜瓦松分配:

X∼Poisson⁡(λ),P(X=k)=e−λλkk!X\sim\operatorname{Poisson}(\lambda),\qquad P(X=k)=\frac{e^{-\lambda}\lambda^k}{k!}

其中 λ\lambda 是該時間區間的平均到達人數。平均到達間隔為 100100 秒,所以每秒平均到達率為 1/1001/100 人;長度為 tt 秒的區間,其平均到達人數為:

λ=t100\lambda=\frac{t}{100}

卜瓦松分配的期望值與變異數皆為 λ\lambda,標準差為 λ\sqrt{\lambda}。

解題方法

先將各選項的時間換算成秒,再用平均到達間隔求出該區間的 λ\lambda,最後依題目要求計算「少於」、「至少」或「恰好」的人數機率。

  • 5 分鐘: 300/100=3300/100=3,所以 λ=3\lambda=3。
  • 10 分鐘: 600/100=6600/100=6,所以 λ=6\lambda=6。
  • 15 分鐘: 900/100=9900/100=9,所以 λ=9\lambda=9。

選項分析

(A) 錯誤。 5 分鐘內少於 2 人,表示到達人數為 0 人或 1 人:

P(X<2)=P(X=0)+P(X=1)=e−3(1+3)≈0.1991P(X<2)=P(X=0)+P(X=1) =e^{-3}\left(1+3\right) \approx 0.1991

題目所列的 0.01990.0199 小數點位置錯誤。

(B) 正確。 5 分鐘內有 4 人或更多,表示 X≥4X\geq 4。用補事件計算:

🔒

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 9 題

An experiment is conducted by rolling three fair dice simultaneously. The pips on dice are colored red on the 1 and 4 sides and colored black on the rest. Let P[Y=x] indicate the probability of a sum of x when rolling three dice. Which of the following descriptions is/are correct?
(A) P[Y=17] = 1/72;
(B) P[Y<5] = 1/54;
(C) P[Y=3] = 1/72;
(D) The probability of getting at least a red side given a sum ≥ 16 is 3/10;
(E) The probability of getting all three red sides given Y ≤ 4 is 1/3.

登入後即可作答並保存紀錄。

這一題的完整詳解

此題考察關於三個骰子同時投擲的機率計算,包含點數和、顏色分佈以及條件機率。

首先,我們需要釐清骰子的設定:

  • 每個骰子有 6 個面。
  • 點數 1 和 4 的面是紅色 (Red)。
  • 點數 2, 3, 5, 6 的面是黑色 (Black)。
  • 每個骰子投擲結果是獨立且均勻的(公平骰子)。
  • 總共有 6×6×6=2166 \times 6 \times 6 = 216 種可能的組合。

計算每個骰子出現紅色或黑色的機率:

  • 點數為 1 或 4 的機率 (紅色):P(Red)=2/6=1/3P(Red) = 2/6 = 1/3。
  • 點數為 2, 3, 5, 6 的機率 (黑色):P(Black)=4/6=2/3P(Black) = 4/6 = 2/3。

我們將用 Y 表示三個骰子點數的總和。

逐一評估選項:

(A) P[Y=17] = 1/72;

  • 三個骰子的最大點數和是 6+6+6=186+6+6=18。
  • 要得到總和 17,可能的組合是 (5, 6, 6) 及其排列。
  • 組合 (5, 6, 6) 的排列有:(5,6,6), (6,5,6), (6,6,5)。共 3 種。
  • 每個組合的機率是 (1/6)×(1/6)×(1/6)=1/216(1/6) \times (1/6) \times (1/6) = 1/216。
  • 所以,P(Y=17)=3×(1/216)=3/216=1/72P(Y=17) = 3 \times (1/216) = 3/216 = 1/72。
  • 此敘述正確。

(B) P[Y<5] = 1/54;

  • Y < 5 表示 Y 的值可以是 3 或 4 (因為最小點數和是 1+1+1=3)。
  • 計算 P(Y=3):
    • 只有一種組合:(1, 1, 1)。
    • P(Y=3)=1×(1/216)=1/216P(Y=3) = 1 \times (1/216) = 1/216。
  • 計算 P(Y=4):
    • 可能的組合是 (1, 1, 2) 及其排列。
    • 組合 (1, 1, 2) 的排列有:(1,1,2), (1,2,1), (2,1,1)。共 3 種。
    • P(Y=4)=3×(1/216)=3/216P(Y=4) = 3 \times (1/216) = 3/216。
  • P(Y<5)=P(Y=3)+P(Y=4)=1/216+3/216=4/216=1/54P(Y < 5) = P(Y=3) + P(Y=4) = 1/216 + 3/216 = 4/216 = 1/54。
  • 此敘述正確。

(C) P[Y=3] = 1/72;

  • 我們在 (B) 中已經計算過:要得到總和 3,只有一種組合 (1, 1, 1)。
  • P(Y=3)=1/216P(Y=3) = 1/216。
  • 題目給的機率是 1/72,這是錯誤的。此敘述錯誤。

(D) The probability of getting at least a red side given a sum ≥ 16 is 3/10;

  • 這是一個條件機率:P(at least one red side∣Y≥16)P(\text{at least one red side} | Y \ge 16)。
  • 首先,找出總和 ≥16\ge 16 的情況。總和最大是 18。
  • Y=18: (6, 6, 6) - 1 種組合。
  • Y=17: (5, 6, 6) 的排列 - 3 種組合 (如 (A) 所述)。
  • Y=16:
    • 組合 (4, 6, 6) 的排列:(4,6,6), (6,4,6), (6,6,4) - 3 種。
    • 組合 (5, 5, 6) 的排列:(5,5,6), (5,6,5), (6,5,5) - 3 種。
🔒

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 10 題

An official institute reported that families in the United States planning to travel out of town over
the Memorial Day weekend spend an average of 749USDollars(USD).Assumethattheamountspentisnormallydistributedwithastandarddeviationof749 US Dollars (USD). Assume that the amount spent is normally distributed with a standard deviation of 225 USD. Which of the
following is/are correct?
(A) The probability of family expenses for the weekend being less than 400is0.0606;(B)Theprobabilityoffamilyexpensesfortheweekendbeing400 is 0.0606; (B) The probability of family expenses for the weekend being 800 or more is 0.5910;
(C) The probability that family expenses for the weekend will be between 500and500 and 1000 is 0.7351;
(D) To be for the 5% of the families with the most expensive travel plans, a
family has to spend $857 or more during the weekend;
(E) None of the above.

登入後即可作答並保存紀錄。

這一題的完整詳解

核心觀念

本題評量常態分配(Normal Distribution)之機率計算與分位數(Quantiles)求解。

設隨機變數 XX 代表美國家庭在國殤紀念日週末出遊的支出金額,根據題意:

X∼N(μ,σ2)=N(749,2252)X \sim N(\mu, \sigma^2) = N(749, 225^2)

其中平均數 μ=749\mu = 749,標準差 σ=225\sigma = 225。

進行常態分配機率計算時,需透過**標準化變數轉換(Z-score Transformation)**轉換為標準常態分配 Z∼N(0,1)Z \sim N(0, 1):

Z=X−μσZ = \frac{X - \mu}{\sigma}

令 Φ(z)=P(Z≤z)\Phi(z) = P(Z \le z) 為標準常態累積分配函數(CDF)。


解題方法與選項分析

(A) 選項 (A) 錯誤

  • 計算目標:P(X<400)P(X < 400)
  • 標準化計算:
Z=400−749225=−349225≈−1.55Z = \frac{400 - 749}{225} = \frac{-349}{225} \approx -1.55
  • 查表求解:
P(X<400)=P(Z<−1.55)=Φ(−1.55)≈0.0606P(X < 400) = P(Z < -1.55) = \Phi(-1.55) \approx 0.0606
  • 辨析:雖然數值計算 P(Z<−1.55)≈0.0606P(Z < -1.55) \approx 0.0606 正確,但本題為研究所入學考試之單選題(若為複選題需另行確認,但通常管研所選擇題若無特別註明多為單選題,惟須逐一精確檢驗各選項是否完全符合)。經精確計算 Z=−1.5511Z = -1.5511,查表得 P(Z<−1.55)=0.0606P(Z < -1.55) = 0.0606,選項 (A) 為正確敘述。

(B) 選項 (B) 錯誤

  • 計算目標:P(X≥800)P(X \ge 800)
  • 標準化計算:
Z=800−749225=51225≈0.23Z = \frac{800 - 749}{225} = \frac{51}{225} \approx 0.23
  • 查表求解:
P(X≥800)=P(Z≥0.23)=1−Φ(0.23)≈1−0.5910=0.4090P(X \ge 800) = P(Z \ge 0.23) = 1 - \Phi(0.23) \approx 1 - 0.5910 = 0.4090
  • 辨析:選項中的 0.5910 實為累積機率 P(X≤800)=P(Z≤0.23)≈0.5910P(X \le 800) = P(Z \le 0.23) \approx 0.5910,題目所求為「800 或更多(800800 or more)」,應為右尾機率 1−0.5910=0.40901 - 0.5910 = 0.4090。此為常見的正負號與尾端方向陷阱。

(C) 選項 (C) 錯誤

  • 計算目標:P(500≤X≤1000)P(500 \le X \le 1000)
  • 標準化計算:
Z1=500−749225=−249225≈−1.11Z_1 = \frac{500 - 749}{225} = \frac{-249}{225} \approx -1.11 Z2=1000−749225=251225≈1.12Z_2 = \frac{1000 - 749}{225} = \frac{251}{225} \approx 1.12
  • 查表求解:
P(500≤X≤1000)=P(−1.11≤Z≤1.12)=Φ(1.12)−Φ(−1.11)P(500 \le X \le 1000) = P(-1.11 \le Z \le 1.12) = \Phi(1.12) - \Phi(-1.11)

查標準常態分配表:

🔒

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 11 題

Assume 30% of emails that are received are tracked using software that can tell the email sender
when, where, and on what type of device the email was opened. Suppose we randomly select 8
received emails. Which of the following is/are correct?
(A) The expected number of these emails that are tracked is 2.4;
(B) The variance of these emails that are tracked is 1.68;
(C) The probability that exactly 3 mails are tracked is 0.2541;
(D) The probability that less than 3 emails are tracked is 0.8059;
(E) None of the above.

登入後即可作答並保存紀錄。

這一題的完整詳解

核心觀念

本題考查二項分配。令 XX 表示 8 封電子郵件中被追蹤的數量。

每封電子郵件被追蹤的機率為 p=0.3p=0.3,未被追蹤的機率為 1−p=0.71-p=0.7,抽取郵件數為 n=8n=8,因此:

X∼Binomial⁡(8,0.3)X\sim\operatorname{Binomial}(8,0.3)

二項分配的重要公式為:

E(X)=npE(X)=np Var⁡(X)=np(1−p)\operatorname{Var}(X)=np(1-p) P(X=k)=(nk)pk(1−p)n−kP(X=k)=\binom{n}{k}p^k(1-p)^{n-k}

解題方法

依序計算期望值、變異數及指定事件的機率,再逐一判斷各選項。

選項分析

(A) 期望被追蹤的電子郵件數為 2.4

E(X)=np=8(0.3)=2.4E(X)=np=8(0.3)=2.4

因此 (A) 正確。


(B) 變異數為 1.68

Var⁡(X)=np(1−p)\operatorname{Var}(X)=np(1-p) =8(0.3)(0.7)=1.68=8(0.3)(0.7)=1.68

因此 (B) 正確。


(C) 恰有 3 封電子郵件被追蹤的機率為 0.2541

P(X=3)=(83)(0.3)3(0.7)5P(X=3)=\binom{8}{3}(0.3)^3(0.7)^5 =56(0.027)(0.16807)=0.2541216≈0.2541=56(0.027)(0.16807) =0.2541216 \approx 0.2541

因此 (C) 正確。


🔒

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 12 題

PPP Marketing Research Company charges to a client on the assumption that telephone surveys
can be completed in a mean time of 15 minutes or less. If a longer mean survey time is necessary,
a premium rate is charged. Based upon past studies, the population standard deviation is assumed
known with σ = 4 minutes. The Company wants to determine whether the premium rate should
apply for a new business. Thus, a sample of 35 pilot surveys was attempted with a mean survey
time of 17 minutes. A hypothesis testing is conducted with the level of significance = 0.01. Which
of the following is/are correct?
(A) The null hypothesis is set as µ₁=15, where μ₁ is the population mean survey time for the new business;
(B) we can conduct z-test and the computed z-score = 0.5;
(C) the critical value for this testing is 2.575;
(D) The p-value is 0.0015;
(E) We should not reject the null hypothesis.

登入後即可作答並保存紀錄。

這一題的完整詳解

核心觀念

本題探討單一母體平均數之右尾假設檢定(One-tailed ZZ-test for Population Mean)。
主要涵蓋觀念與公式如下:

  1. 假設檢定的建立:研究者欲檢定新業務是否需要收取加成費率(即平均訪問時間是否顯著大於 15 分鐘),對立假設設為欲證明的方向,虛無假設則為現行常態基準。
  2. 檢定統計量(Test Statistic):當母體標準差 σ\sigma 已知,且樣本數大(n=35≥30n=35 \ge 30),根據中央極限定理(CLT),樣本平均數 Xˉ\bar{X} 的抽樣分配服從常態分配: Z=Xˉ−μ0σ/n∼N(0,1)Z = \frac{\bar{X} - \mu_0}{\sigma / \sqrt{n}} \sim N(0, 1)
  3. 臨界值法則與 pp 值法則:在顯著水準 α\alpha 下,單尾(右尾)檢定的拒絕域為 Z>zαZ > z_{\alpha},或當 p-value≤αp\text{-value} \le \alpha 時拒絕虛無假設 H0H_0。

解題方法

步驟一:摘要題目已知條件

  • 母體平均數基準值:μ0=15\mu_0 = 15
  • 母體標準差(已知):σ=4\sigma = 4
  • 試訪樣本數:n=35n = 35
  • 樣本平均時間:Xˉ=17\bar{X} = 17
  • 顯著水準:α=0.01\alpha = 0.01

步驟二:建立假說

市調公司現行的收費基準是「平均時間在 15 分鐘以內」;若時間超過 15 分鐘則加收費用。公司欲檢驗新業務是否應適用加成費率,因此欲證明的研究假說為 μ>15\mu > 15:

H0:μ≤15(或記為 μ=15)H_0: \mu \le 15 \quad (\text{或記為 } \mu = 15) H1:μ>15H_1: \mu > 15

此為單尾(右尾)檢定。

步驟三:計算檢定統計量之數值

因母體標準差 σ\sigma 已知,採用 ZZ 檢定:

Z=Xˉ−μ0σ/n=17−154/35=24/5.9161=20.6761≈2.958Z = \frac{\bar{X} - \mu_0}{\sigma / \sqrt{n}} = \frac{17 - 15}{4 / \sqrt{35}} = \frac{2}{4 / 5.9161} = \frac{2}{0.6761} \approx 2.958

步驟四:確定臨界值(Critical Value)與拒絕域

在顯著水準 α=0.01\alpha = 0.01 之右尾檢定下:

  • 臨界值為 z0.01≈2.326z_{0.01} \approx 2.326(若為雙尾檢定,臨界值才是 z0.005≈2.576z_{0.005} \approx 2.576)。
  • 拒絕域為 Z>2.326Z > 2.326。

步驟五:計算 pp 值(pp-value)與做成決策

  • 由標準常態分配表查得: p-value=P(Z≥2.96)=1−P(Z<2.96)≈1−0.9985=0.0015p\text{-value} = P(Z \ge 2.96) = 1 - P(Z < 2.96) \approx 1 - 0.9985 = 0.0015
  • 檢定判斷:
🔒

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 13 題

Which of the following descriptions about Type I error rate (a) and Type II error rate (β) is/are correct?
(A) Rejecting a null hypothesis when the alternative hypothesis is true is called Type I error;
(B) Rejecting a null hypothesis when the null hypothesis is false is called Type II error;
(C) The power of a statistical test is measured as (1- β);
(D) Increasing a will decrease β;
(E) For a given level of a, an increase in the sample size will decrease β.

登入後即可作答並保存紀錄。

這一題的完整詳解

此題考察假設檢定中的兩類錯誤 (Type I error, Type II error) 和檢定效能 (Power of a test)。

我們定義:

  • H0H_0: 虛無假設 (Null hypothesis)
  • H1H_1: 對立假設 (Alternative hypothesis)

兩種錯誤及其機率:

  • Type I error (第一類錯誤):當虛無假設 H0H_0 為真時,卻拒絕了 H0H_0。其機率用 α\alpha 表示。
  • Type II error (第二類錯誤):當虛無假設 H0H_0 為偽(即對立假設 H1H_1 為真)時,卻未能拒絕 H0H_0。其機率用 β\beta 表示。

檢定效能 (Power of a test):指當對立假設 H1H_1 為真時,正確地拒絕虛無假設 H0H_0 的機率。其值為 1−β1 - \beta。

現在逐一評估選項:

(A) Rejecting a null hypothesis when the alternative hypothesis is true is called Type I error;

  • Type I error 是指「當 H0H_0 為真時,拒絕 H0H_0」。
  • 當 H1H_1 為真時,代表 H0H_0 為偽。
  • 所以,這個敘述描述的是「當 H0H_0 為偽時,拒絕 H0H_0」。這是正確地拒絕 H0H_0,也就是檢定效能,而不是 Type I error。
  • 此敘述錯誤。

(B) Rejecting a null hypothesis when the null hypothesis is false is called Type II error;

  • Type II error 是指「當 H0H_0 為偽時,未能拒絕 H0H_0」。
🔒

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 14 題

The time that married men with children spend on child care averages 6.4 hours per week. A
researcher would like to determine if time married men in a specified local area spend on child
care per week differs from the reported mean of 6.4 hours per week. A sample of 25 married
couples was used with the data collected showing the hours per week the husband spends on child
with the mean of 7.0 hours and the standard deviation of 1.5. Which of the following calculation
is/are correct?
(A) The null hypothesis is set as μ₁=7.0, where μ₁ is the population mean hours
spent on child care per week for married men in the specified local area;
(B) The 95% confidence interval (CI) is 7.0±(2.0595)(0.3);
(C) The 99% CI is 6.4±(2.7969)(0.3);
(D) The p-value of the statistical testing is between 0.025~0.05;
(E) At a significance level of 0.05, do not reject the null hypothesis.

登入後即可作答並保存紀錄。

這一題的完整詳解

核心觀念

本題考查單一母體平均數的假設檢定與信賴區間。母體標準差未知,且樣本數為 n=25n=25,因此使用單一樣本 tt 檢定。

題目要判斷指定地區已婚男性每週照顧小孩的平均時間是否「不同於」6.46.4 小時,屬於雙尾檢定。

令 μ\mu 為指定地區已婚男性每週照顧小孩的母體平均時間,則:

H0:μ=6.4H_0:\mu=6.4 H1:μ≠6.4H_1:\mu\ne 6.4

已知:

n=25,xˉ=7.0,s=1.5n=25,\qquad \bar{x}=7.0,\qquad s=1.5

樣本平均數的標準誤為:

SE=sn=1.525=1.55=0.3SE=\frac{s}{\sqrt{n}} =\frac{1.5}{\sqrt{25}} =\frac{1.5}{5} =0.3

自由度為:

df=n−1=24df=n-1=24

解題方法

1. 計算檢定統計量

單一樣本 tt 統計量為:

t=xˉ−μ0s/nt=\frac{\bar{x}-\mu_0}{s/\sqrt{n}}

其中 μ0=6.4\mu_0=6.4,因此:

t=7.0−6.40.3=0.60.3=2.00t=\frac{7.0-6.4}{0.3} =\frac{0.6}{0.3} =2.00

由於是雙尾檢定,pp 值為:

p=2P(T24≥2.00)≈0.057p=2P(T_{24}\ge 2.00)\approx 0.057

因為:

0.057>0.050.057>0.05

所以在顯著水準 0.050.05 下,不拒絕虛無假設。

2. 建立 95% 信賴區間

單一母體平均數的 95% 信賴區間為:

xˉ±t0.025,24sn\bar{x}\pm t_{0.025,24}\frac{s}{\sqrt{n}}

查 tt 分配表:

t0.025,24≈2.0639t_{0.025,24}\approx 2.0639

因此:

95% CI=7.0±(2.0639)(0.3)95\%\ CI =7.0\pm(2.0639)(0.3)

即:

95% CI=7.0±0.6192=(6.3808, 7.6192)95\%\ CI =7.0\pm0.6192 =(6.3808,\ 7.6192)

3. 建立 99% 信賴區間

99% 信賴區間使用:

t0.005,24≈2.7969t_{0.005,24}\approx 2.7969

因此:

99% CI=7.0±(2.7969)(0.3)99\%\ CI =7.0\pm(2.7969)(0.3)

即:

🔒

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 15 題

A market research firm used a sample of individuals to rate the purchase potential of a particular
new product before and after the individuals saw a new TV advertisement about the product. The
purchase potential ratings were based on a 0 to 10 scale, with higher values indicating a higher
purchase potential. The paired two-sample test for the means was then conducted to detect
whether the mean rating "after" would be different from the mean rating "before." (Difference
rating after - rating before) Eight individuals were surveyed, and the result is as follows:
Purchase Rating
Individual
Before
After
1
6
6
2
5
7
3
8
9
4
5
3
5
3
4
6
7
7
7
4
6
8
5
6
9
2
5
Which of the following calculation is/are correct?
(A) The null hypothesis is μ₁ = μ₂; where μι the mean population rating before while µ₂ = the mean population rating after;
(B) The mean difference = 0.8889;
(C) The t-statistic is 1.8353;
(D) The critical value is 1.8595 at the significance level of 0.05;
(E) The p-value is between 0.1 ~ 0.2.

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「成對樣本平均數差異的 tt 檢定」(paired-sample tt test)。

同一批受試者分別接受「廣告前」與「廣告後」評分,因此兩組資料彼此相關,不能使用獨立樣本 tt 檢定。應先對每位受試者計算差異:

di=Afteri−Beforeid_i=\text{After}_i-\text{Before}_i

檢定假設為:

H0:μd=0H_0:\mu_d=0 H1:μd≠0H_1:\mu_d\ne 0

其中 μd\mu_d 為母體平均差異。因為題目使用 “different from”,所以是雙尾檢定。

題幹文字寫有八位受試者,但表格實際列出九組資料;且選項 B、C 的數值均以九組資料計算。以下依表格中的九組資料作答。


解題方法與計算

1. 計算各組差異

個人BeforeAfterdid_i
1660
2572
3891
453-2
5341
6770
7462
8561
9253

因此:

∑di=0+2+1−2+1+0+2+1+3=8\sum d_i=0+2+1-2+1+0+2+1+3=8

樣本數為 n=9n=9,故平均差異為:

dˉ=∑din=89=0.8889\bar d=\frac{\sum d_i}{n} =\frac{8}{9} =0.8889

2. 計算差異的樣本標準差

先計算平方和:

∑di2=02+22+12+(−2)2+12+02+22+12+32=24\sum d_i^2 =0^2+2^2+1^2+(-2)^2+1^2+0^2+2^2+1^2+3^2 =24

差異值的離均差平方和為:

∑(di−dˉ)2=∑di2−(∑di)2n\sum(d_i-\bar d)^2 =\sum d_i^2-\frac{(\sum d_i)^2}{n} =24−829=24−649=1529=16.8889=24-\frac{8^2}{9} =24-\frac{64}{9} =\frac{152}{9} =16.8889

樣本變異數:

sd2=∑(di−dˉ)2n−1=16.88898=2.1111s_d^2=\frac{\sum(d_i-\bar d)^2}{n-1} =\frac{16.8889}{8} =2.1111

樣本標準差:

sd=2.1111=1.45297s_d=\sqrt{2.1111}=1.45297

3. 計算 tt 統計量

成對樣本 tt 統計量為:

t=dˉ−μd,0sd/nt=\frac{\bar d-\mu_{d,0}}{s_d/\sqrt n}

在虛無假設下 μd,0=0\mu_{d,0}=0,所以:

🔒

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 16 題

The sample data below represent the number of late arrival and on-time flights for three airlines.
You want to detect whether the variable of "airline company" is independent from the risk of late
arrival. Assume the level of significance is set at 0.05.
Airline
Flight Status
Airline A
Airline B
Airline C
Late Arrival
39
51
60
On Time
261
249
340
Which of the following is/are correct?
(A) The null hypothesis is that the variable of "airline company" is not independent from the risk of late arrival;
(B) The test statistic is 1.8824;
(C) The degree of freedom is 5 for the statistical testing;
(D) The critical value is 7.3778;
(E) The p-value is between 0.05~0.1.

登入後即可作答並保存紀錄。

這一題的完整詳解

本題考查的核心觀念為雙因子卡方獨立性檢定(Chi-Square Test for Independence)。此檢定旨在評估兩個名目/類別變數(航空公司與班機延誤風險)之間是否存在統計上的關聯性,解題關鍵包含:虛無假設與對立假設的設定規範、期望次數(Expected Frequencies)與卡方檢定統計量(Test Statistic)的計算、自由度(Degrees of Freedom, dfdf)的判定,以及臨界值(Critical Value)與 pp 值(pp-value)的判讀。


詳細解題過程與步驟說明

步驟一:建立假設檢定(Hypothesis Testing)

在卡方獨立性檢定中,虛無假設(H0H_0)必須設定為兩變數相互獨立(即無關聯),而對立假設(H1H_1)則為兩變數不獨立(即有關聯):

  • H0H_0:航空公司與班機延誤風險相互獨立("Airline company" is independent from the risk of late arrival)。
  • H1H_1:航空公司與班機延誤風險不獨立("Airline company" is not independent from the risk of late arrival)。

原理說明:獨立性檢定預設「兩變數沒有關聯」為基準情況(H0H_0),檢定目的是在尋找足夠的統計證據來拒絕該無關聯的假設。選項 (A) 將虛無假設寫為「不獨立」,觀念完全顛倒。


步驟二:計算各邊際總和與期望次數(Expected Frequencies)

將題目給予的交叉表(Contingency Table)計算邊際總和:

班機狀態 \ 航空公司Airline AAirline BAirline C列總和(RiR_i)
Late Arrival395160150
On Time261249340850
行總和(CjC_j)300300400總樣本數 N=1000N = 1000

依據獨立性假設,儲存格 (i,j)(i, j) 的期望次數公式為:
Eij=Ri×CjNE_{ij} = \frac{R_i \times C_j}{N}

原理說明:若兩變數獨立,則條件機率相乘即為聯合機率,因此每一個儲存格在獨立假設下的理論期望次數,等於該列總和乘上該行總和再除以總樣本數。

計算各儲存格之期望次數 EijE_{ij}:

  • E11=150×3001000=45E_{11} = \frac{150 \times 300}{1000} = 45
  • E12=150×3001000=45E_{12} = \frac{150 \times 300}{1000} = 45
  • E13=150×4001000=60E_{13} = \frac{150 \times 400}{1000} = 60
  • E21=850×3001000=255E_{21} = \frac{850 \times 300}{1000} = 255
  • E22=850×3001000=255E_{22} = \frac{850 \times 300}{1000} = 255
  • E23=850×4001000=340E_{23} = \frac{850 \times 400}{1000} = 340

步驟三:計算卡方檢定統計量(Test Statistic)

卡方檢定統計量公式為:
χ2=∑i=1r∑j=1c(Oij−Eij)2Eij\chi^2 = \sum_{i=1}^{r}\sum_{j=1}^{c} \frac{(O_{ij} - E_{ij})^2}{E_{ij}}

將各儲存格的觀察次數 OijO_{ij} 與期望次數 EijE_{ij} 代入:

🔒

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 17 題

Suppose you want to investigate the association between two numerical variables. Which of the following is/are appropriate?
(A) Conduct a line chart to visualize their association;
(B) Estimate a correlation coefficient, which ranges from -1 to +1;
(C) Conduct a t-test to compare whether these two variables are statistically different;
(D) Develop a simple linear regression model and detect whether the slope is different from 0;
(E) Conduct a Chi-square test to detect whether these two variables are independent or not.

登入後即可作答並保存紀錄。

這一題的完整詳解

此題考察如何研究兩個數值變數 (numerical variables) 之間的關聯性。

(A) Line chart(線圖)通常用於顯示時間序列數據的趨勢,或隨時間變化的數據。它不適合用來直接視覺化兩個數值變數之間的關聯性。要視覺化兩個數值變數的關聯性,通常使用散佈圖 (scatter plot)。因此,(A) 不適合。

(B) Correlation coefficient(相關係數)是一種衡量兩個數值變數之間線性關聯強度和方向的統計量,其值介於 -1 到 +1 之間。相關係數是研究兩個數值變數關聯性的標準方法之一。因此,(B) 是適當的。

(C) t-test(t 檢定)通常用於比較一個樣本平均數與已知母體平均數的差異,或比較兩個獨立樣本或配對樣本的平均數差異。

🔒

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 18 題

Which of the following is/are correct about ANOVA?
(A) One assumption of ANOVA is that, for each population, the response variable is normally distributed;
(B) One assumption of ANOVA is that the variance of the response variable is the same for all of the populations;
(C) Suppose three groups are compared, the null hypothesis is µ₁= µ₂= µ₃, while the alternative hypothesis is µ₁≠ µ₂≠ µ₃;
(D) One-way ANOVA is a one-tailed test, while two-way ANOVA is a two-tailed test;
(E) If the test result is not to reject the null hypothesis, we must perform a post-hoc test to confirm that any two groups are truly no different from each other.

登入後即可作答並保存紀錄。

這一題的完整詳解

此題考察變異數分析 (ANOVA) 的基本假設、假設檢定方式及後續分析。

ANOVA 的主要假設包括:

  1. 獨立性 (Independence):各組的觀測值是相互獨立的。
  2. 常態性 (Normality):各組的母體應服從常態分佈。
  3. 同質性 (Homogeneity of variances):各組母體的變異數相等(同質變異數)。

現在逐一評估選項:

(A) One assumption of ANOVA is that, for each population, the response variable is normally distributed;

  • 這是 ANOVA 的常態性假設。此敘述正確。

(B) One assumption of ANOVA is that the variance of the response variable is the same for all of the populations;

  • 這是 ANOVA 的同質變異數假設。此敘述正確。

(C) Suppose three groups are compared, the null hypothesis is µ₁= µ₂= µ₃, while the alternative hypothesis is µ₁≠ µ₂≠ µ₃;

  • 在 ANOVA 中,虛無假設 H0H_0 是所有組別的母體平均數都相等,即 μ1=μ2=⋯=μk\mu_1 = \mu_2 = \dots = \mu_k。
  • 對立假設 H1H_1 是至少有一個組別的母體平均數不同,即「不全相等」。
🔒

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 19 題

The following ANOVA table was the statistical results for a completely randomized design among
three treatment groups, assuming the significance level = 0.05:
ANOVA
Source of
Variation
SS
df
MS
F
P-value
F critical
value
Between Groups
1488
?
①
3
?
Within Groups
2030
?
②
Total
?
17
Which of the following descriptions is/are correct?
(A) ① = 744;
(B) ② = 135.3333;
(C) ③ = 5.4975;
(D) ④ = 3.3439;
(E) There is not enough information to complete all of the calculations.

登入後即可作答並保存紀錄。

這一題的完整詳解

此題考察變異數分析 (ANOVA) 表格的填寫與計算,以及 F 檢定的相關概念。

ANOVA 表格結構:

Source of VariationSS (Sum of Squares)df (Degrees of Freedom)MS (Mean Square)FP-valueF critical value
Between GroupsSS_betweendf_betweenMS_betweenFP-valueF critical value
Within GroupsSS_withindf_withinMS_within
TotalSS_totaldf_total

已知關係:

  • MS=SS/dfMS = SS / df
  • F=MSbetween/MSwithinF = MS_{between} / MS_{within}
  • SStotal=SSbetween+SSwithinSS_{total} = SS_{between} + SS_{within}
  • dftotal=dfbetween+dfwithindf_{total} = df_{between} + df_{within}

根據題目給予的 ANOVA 表格資訊:

  • Three treatment groups,表示組別數 (k) = 3。

  • dfbetween=k−1=3−1=2df_{between} = k - 1 = 3 - 1 = 2。

  • dftotal=17df_{total} = 17。

  • dftotal=dfbetween+dfwithindf_{total} = df_{between} + df_{within},所以 17=2+dfwithin17 = 2 + df_{within},則 dfwithin=17−2=15df_{within} = 17 - 2 = 15。

  • SSbetween=1488SS_{between} = 1488

  • SSwithin=2030SS_{within} = 2030

  • SStotal=SSbetween+SSwithin=1488+2030=3518SS_{total} = SS_{between} + SS_{within} = 1488 + 2030 = 3518。

🔒

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 20 題

Below is the estimated linear regression equation based on 10 observations, where x₁ and x₂ are
numerical variables:
ŷ = 28+0.6x1 + 0.5x2
where SST (total sum of squares) =6725, SSR (regression sum of squares) =6215, sb1=0.08, and
sb2=0.06. The b₁ and b₂ are the estimated values of β₁ and β₂ for the variable x1, x2, respectively; sb1,
sb2 are the standard errors of the estimated coefficients. Which of the following is/are correct?
(A) R-square=6215/6725;
(B) MSR (regression mean square) = 6215/2;
(C) MSE (mean square error) = 510/7;
(D) Given a=0.05, t-test with the critical values of ±2.3646 can be used to detect whether β₁
is significantly different from zero;
(E) β₂ is not significantly different from zero given a=0.05.

登入後即可作答並保存紀錄。

這一題的完整詳解

此題考察複迴歸分析 (Multiple Linear Regression) 的基本概念和計算。

已知資訊:

  • 迴歸方程式:y^=28+0.6x1+0.5x2\hat{y} = 28 + 0.6x_1 + 0.5x_2。
  • 樣本數 n=10n = 10。
  • SST=6725SST = 6725 (總平方和)。
  • SSR=6215SSR = 6215 (迴歸平方和)。
  • b1=0.6b_1 = 0.6 (x₁ 的迴歸係數估計值)。
  • b2=0.5b_2 = 0.5 (x₂ 的迴歸係數估計值)。
  • sb1=0.08sb_1 = 0.08 (x₁ 的迴歸係數標準誤)。
  • sb2=0.06sb_2 = 0.06 (x₂ 的迴歸係數標準誤)。
  • 顯著水準 α=0.05\alpha = 0.05。

首先,我們需要計算一些基本統計量:

  • SSESSE (殘差平方和) =SST−SSR=6725−6215=510= SST - SSR = 6725 - 6215 = 510。
  • dfregression=自變數個數=2df_{regression} = \text{自變數個數} = 2 (因為有 x1x_1 和 x2x_2)。
  • dfresidual=n−(自變數個數+1)=10−(2+1)=10−3=7df_{residual} = n - (\text{自變數個數} + 1) = 10 - (2 + 1) = 10 - 3 = 7。
  • dftotal=n−1=10−1=9df_{total} = n - 1 = 10 - 1 = 9。

現在逐一評估選項:

(A) R-square=6215/6725;

  • R2R^2 (決定係數) 的計算公式為 R2=SSR/SSTR^2 = SSR / SST。
  • R2=6215/6725R^2 = 6215 / 6725。
  • 此敘述正確。

(B) MSR (regression mean square) = 6215/2;

  • MSR=SSR/dfregression=6215/2MSR = SSR / df_{regression} = 6215 / 2。
  • 此敘述正確。

(C) MSE (mean square error) = 510/7;

  • MSE=SSE/dfresidual=510/7MSE = SSE / df_{residual} = 510 / 7。
  • 此敘述正確。

(D) Given a=0.05, t-test with the critical values of ±2.3646 can be used to detect whet

🔒

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