111 年 國立政治大學統計學系碩士班《統計方法》

📄 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12 分

A Center of Sleep Disorder conducted a survey to determine whether the hours of sleep per night are independent of age. A random sample of 500 individuals was asked two questions: (i) hours of sleep per night (with four options: fewer than 6, 6 to 6.9, 7 to 7.9, and 8 or more); (ii) age (younger than 30, and 30 or older). The data are given in the following table.

🖼️【此處有附圖,請對照原卷】

Age

Hours of sleepYounger than 3030 or older
fewer than 66090
6 to 6.97070
7 to 7.96060
8 or more5040

(a) Using a 0.05 significance level, conduct a hypothesis test to determine whether the hours of sleep per night are independent of age. Please complete the hypothesis test. (10%)
(b) What is your interpretation about the result in (a)? (2%)

🖼️ 本題含圖表,以下為原卷對應頁面:
原卷第 1 頁

登入後即可作答並保存紀錄。

這一題的完整詳解

此題考驗獨立性檢定(Test of Independence)。

(a) 獨立性檢定步驟:

  1. 建立虛無與對立假設:

    • H0H_0:睡眠時數與年齡是獨立的。
    • H1H_1:睡眠時數與年齡不是獨立的。
  2. 設定顯著水準:

    • α=0.05\alpha = 0.05。
  3. 計算期望次數 (Expected Frequencies):
    首先,計算各行(睡眠時數)和各列(年齡)的總次數,以及總樣本數。

    • 總樣本數 N=500N = 500。
    • 各睡眠時數總計:
      • fewer than 6: 60+90=15060 + 90 = 150
      • 6 to 6.9: 70+70=14070 + 70 = 140
      • 7 to 7.9: 60+60=12060 + 60 = 120
      • 8 or more: 50+40=9050 + 40 = 90
    • 各年齡總計:
      • Younger than 30: 60+70+60+50=24060 + 70 + 60 + 50 = 240
      • 30 or older: 90+70+60+40=26090 + 70 + 60 + 40 = 260

    期望次數 Eij=(Row i total)×(Column j total)NE_{ij} = \frac{(\text{Row } i \text{ total}) \times (\text{Column } j \text{ total})}{N}

    • E11E_{11} (fewer than 6, Younger than 30) = 150×240500=72\frac{150 \times 240}{500} = 72
    • E12E_{12} (fewer than 6, 30 or older) = 150×260500=78\frac{150 \times 260}{500} = 78
    • E21E_{21} (6 to 6.9, Younger than 30) = 140×240500=67.2\frac{140 \times 240}{500} = 67.2
    • E22E_{22} (6 to 6.9, 30 or older) = 140×260500=72.8\frac{140 \times 260}{500} = 72.8
    • E31E_{31} (7 to 7.9, Younger than 30) = 120×240500=57.6\frac{120 \times 240}{500} = 57.6
    • E32E_{32} (7 to 7.9, 30 or older) = 120×260500=62.4\frac{120 \times 260}{500} = 62.4
    • E41E_{41} (8 or more, Younger than 30) = 90×240500=43.2\frac{90 \times 240}{500} = 43.2
    • E42E_{42} (8 or more, 30 or older) = 90×260500=46.8\frac{90 \times 260}{500} = 46.8
  4. 計算檢定統計量 (Test Statistic):
    卡方檢定統計量 χ2=∑i,j(Oij−Eij)2Eij\chi^2 = \sum_{i,j} \frac{(O_{ij} - E_{ij})^2}{E_{ij}}

🔒

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20 分

For testing H0:μ≤22H_0: \mu \le 22 v.s. Hα:μ>22H_\alpha: \mu > 22, a random sample of 256 individuals is selected from the population. The population variance is 400. The power is 0.2005, denoted by 1−β23.41 - \beta_{23.4}, when the actual population mean is 23.4.

(a) What is the significance level adopted in the test? (12%)
(b) In such a hypothesis testing with a sample of 256 individuals, is it possible that the probability of type II error β23.4\beta_{23.4} and the significance level decrease simultaneously? Please give your reason. (8%)

登入後即可作答並保存紀錄。

這一題的完整詳解

核心觀念

本題考查母體變異數已知時的單一母體平均數右尾 ZZ 檢定,以及顯著水準與第二類錯誤的關係。

已知:

  • H0:μ≤22H_0:\mu\le 22
  • Hα:μ>22H_\alpha:\mu>22
  • n=256n=256
  • σ2=400\sigma^2=400,故 σ=20\sigma=20
  • 標準誤為
σXˉ=σn=20256=2016=1\sigma_{\bar X}=\frac{\sigma}{\sqrt n} =\frac{20}{\sqrt{256}} =\frac{20}{16}=1

右尾檢定的拒絕域可寫為:

Xˉ>c\bar X>c

其中 cc 為臨界值。

顯著水準為虛無假設邊界 μ=22\mu=22 下的拒絕機率:

α=P(Xˉ>c∣μ=22)\alpha=P(\bar X>c\mid \mu=22)

當實際平均數為 μ=23.4\mu=23.4 時,檢定力為:

1−β23.4=P(Xˉ>c∣μ=23.4)=0.20051-\beta_{23.4} =P(\bar X>c\mid \mu=23.4) =0.2005

解題方法

(a)求顯著水準

由題意:

P(Xˉ>c∣μ=23.4)=0.2005P(\bar X>c\mid \mu=23.4)=0.2005

因為 Xˉ∼N(23.4,12)\bar X\sim N(23.4,1^2),標準化得:

P(Z>c−23.41)=0.2005P\left( Z>\frac{c-23.4}{1} \right)=0.2005

標準常態分配中:

P(Z>0.84)=0.2005P(Z>0.84)=0.2005

因此:

c−23.41=0.84\frac{c-23.4}{1}=0.84

得到臨界值:

c=23.4+0.84=24.24c=23.4+0.84=24.24

接著在虛無假設的邊界 μ=22\mu=22 下計算顯著水準:

α=P(Xˉ>24.24∣μ=22)\alpha=P(\bar X>24.24\mid \mu=22)

此時 Xˉ∼N(22,12)\bar X\sim N(22,1^2),所以:

α=P(Z>24.24−221)=P(Z>2.24)\alpha =P\left( Z>\frac{24.24-22}{1} \right) =P(Z>2.24)

查標準常態分配表:

P(Z>2.24)=1−Φ(2.24)=1−0.9875=0.0125P(Z>2.24)=1-\Phi(2.24)=1-0.9875=0.0125

故:

α=0.0125\boxed{\alpha=0.0125}

亦即顯著水準為 1.25%1.25\%。


(b)顯著水準與第二類錯誤能否同時下降?

🔒

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35 分

A realty company sells apartments in the Midwest of the United States. Since their customers are interested in the fee they have to pay for heating in winter, a research group in the company collected data and tried to build regression models to predict the heating cost (abbreviated to Heat.cost; denoted by Y). There were three independent variables: the mean temperature in winter (Temp, X1X_1), the centimeters of insulation in the attic (Insu, X2X_2), and the age (years) of the heater (Age, X3X_3). The research group considered the following regression analyses:

Model 1: Y=α+β1X1+ηY = \alpha + \beta_1 X_1 + \eta
Model 2: Y=α+β3X3+δY = \alpha + \beta_3 X_3 + \delta
Model 3: Y=α+β1X1+β2X2+β3X3+ϵY = \alpha + \beta_1 X_1 + \beta_2 X_2 + \beta_3 X_3 + \epsilon

where η\eta, δ\delta, and ϵ\epsilon are error terms. They collected data from 20 of their former customers for data analysis. First of all, the Model 1 is fitted based on the 10 elements given in the following table, and please answer Questions (a) and (b).

CustomerYX₁
1250-5
2360-9
3165-5
4439
59211
6200-8
7355-19
8290-21
9230-13
101206

(a) Find the least squares estimates of α\alpha and β1\beta_1 under Model 1. (12%)
(b) Compute the coefficient of determination, R2R^2, for Model 1. (12%)

An output from a computer software for fitting Model 3 to the data is displayed in the followings. Using the output to answer Questions (c) and (d).

Coefficients:
| | Estimate | Std. Error | t value | Pr(>|t|) |
|-----------|----------|------------|---------|----------|
| (Intercept)| 228.833 | 40.547 | 5.644 | 3.66e-05 *** |
| Temp | -8.348 | 1.361 | -6.134 | 1.44e-05 *** |
| Insu | -6.098 | 1.872 | -3.258 | 0.00494 ** |
| Age | 5.751 | 3.937 | 1.461 | 0.16349 |

Signif. codes: 0 *** 0.001 ** 0.01 * 0.05 '.' 0.1 ' ' 1

Residual standard error: 49.98 on 16 degrees of freedom
Multiple R-squared: 0.8123, Adjusted R-squared: 0.7771
F-statistic: 23.08 on 3 and 16 DF, p-value: 4.695e-06

The output for fitting Model 2 to the data is displayed in the followings.
Coefficients:
| | Estimate | Std. Error | t value | Pr(>|t|) |
|-----------|----------|------------|---------|----------|
| (Intercept)| 86.184 | 48.656 | 1.771 | 0.0934 . |
| Age | 17.009 | 6.303 | 2.699 | 0.0147 * |

(c) What is your interpretation about the result of the F-test for Model 3? (5%)
(d) Based on the fittings, the β3\beta_3 under Model 2 is significant while the β3\beta_3 under Model 3 is not, at a 0.05 significance level. Comment on the phenomenon (Is there any contradiction, or ...?). (6%)

登入後即可作答並保存紀錄。

這一題的完整詳解

此題考驗簡單線性迴歸、複迴歸的參數估計、判定係數,以及複迴歸的 F 檢定與變數篩選的解讀。

Model 1:簡單線性迴歸 Y=α+β1X1+ηY = \alpha + \beta_1 X_1 + \eta

(a) 計算 α\alpha 和 β1\beta_1 的最小平方法估計值:
對於簡單線性迴歸 Y=α+β1X1+ηY = \alpha + \beta_1 X_1 + \eta,最小平方法估計量為:
β^1=∑i=1n(Xi−Xˉ)(Yi−Yˉ)∑i=1n(Xi−Xˉ)2=SXYSXX\hat{\beta}_1 = \frac{\sum_{i=1}^{n}(X_{i} - \bar{X})(Y_{i} - \bar{Y})}{\sum_{i=1}^{n}(X_{i} - \bar{X})^2} = \frac{S_{XY}}{S_{XX}}
α^=Yˉ−β^1Xˉ\hat{\alpha} = \bar{Y} - \hat{\beta}_1 \bar{X}

首先,計算所需的總和:
n=10n = 10
∑Xi=−5−9−5+9+11−8−19−21−13+6=−59\sum X_i = -5 - 9 - 5 + 9 + 11 - 8 - 19 - 21 - 13 + 6 = -59
Xˉ=−5910=−5.9\bar{X} = \frac{-59}{10} = -5.9

∑Yi=250+360+165+43+92+200+355+290+230+120=2005\sum Y_i = 250 + 360 + 165 + 43 + 92 + 200 + 355 + 290 + 230 + 120 = 2005
Yˉ=200510=200.5\bar{Y} = \frac{2005}{10} = 200.5

計算 SXY=∑XiYi−nXˉYˉS_{XY} = \sum X_i Y_i - n \bar{X} \bar{Y}:
∑XiYi=(−5)(250)+(−9)(360)+(−5)(165)+(9)(43)+(11)(92)+(−8)(200)+(−19)(355)+(−21)(290)+(−13)(230)+(6)(120)\sum X_i Y_i = (-5)(250) + (-9)(360) + (-5)(165) + (9)(43) + (11)(92) + (-8)(200) + (-19)(355) + (-21)(290) + (-13)(230) + (6)(120)
∑XiYi=−1250−3240−825+387+1012−1600−6745−6090−2990+720=−13526\sum X_i Y_i = -1250 - 3240 - 825 + 387 + 1012 - 1600 - 6745 - 6090 - 2990 + 720 = -13526
SXY=−13526−10×(−5.9)×(200.5)=−13526+11829.5=−1696.5S_{XY} = -13526 - 10 \times (-5.9) \times (200.5) = -13526 + 11829.5 = -1696.5

計算 SXX=∑Xi2−nXˉ2S_{XX} = \sum X_i^2 - n \bar{X}^2:
∑Xi2=(−5)2+(−9)2+(−5)2+92+112+(−8)2+(−19)2+(−21)2+(−13)2+62\sum X_i^2 = (-5)^2 + (-9)^2 + (-5)^2 + 9^2 + 11^2 + (-8)^2 + (-19)^2 + (-21)^2 + (-13)^2 + 6^2
∑Xi2=25+81+25+81+121+64+361+441+169+36=1304\sum X_i^2 = 25 + 81 + 25 + 81 + 121 + 64 + 361 + 441 + 169 + 36 = 1304
SXX=1304−10×(−5.9)2=1304−10×34.81=1304−348.1=955.9S_{XX} = 1304 - 10 \times (-5.9)^2 = 1304 - 10 \times 34.81 = 1304 - 348.1 = 955.9

β^1=SXYSXX=−1696.5955.9≈−1.7747\hat{\beta}_1 = \frac{S_{XY}}{S_{XX}} = \frac{-1696.5}{955.9} \approx -1.7747

α^=Yˉ−β^1Xˉ=200.5−(−1.7747)×(−5.9)=200.5−10.47073≈190.0293\hat{\alpha} = \bar{Y} - \hat{\beta}_1 \bar{X} = 200.5 - (-1.7747) \times (-5.9) = 200.5 - 10.47073 \approx 190.0293

【答案】α^≈190.0293\hat{\alpha} \approx 190.0293, β^1≈−1.7747\hat{\beta}_1 \approx -1.7747。

(b) 計算判定係數 R2R^2:
R2=SXY2SXXSYYR^2 = \frac{S_{XY}^2}{S_{XX} S_{YY}},其中 SYY=∑Yi2−nYˉ2S_{YY} = \sum Y_i^2 - n \bar{Y}^2。
首先計算 SYYS_{YY}:
∑Yi2=2502+3602+1652+432+922+2002+3552+2902+2302+1202\sum Y_i^2 = 250^2 + 360^2 + 165^2 + 43^2 + 92^2 + 200^2 + 355^2 + 290^2 + 230^2 + 120^2
∑Yi2=62500+129600+27225+1849+8464+40000+126025+84100+52900+14400=547063\sum Y_i^2 = 62500 + 129600 + 27225 + 1849 + 8464 + 40000 + 126025 + 84100 + 52900 + 14400 = 547063
SYY=547063−10×(200.5)2=547063−10×40200.25=547063−402002.5=145060.5S_{YY} = 547063 - 10 \times (200.5)^2 = 547063 - 10 \times 40200.25 = 547063 - 402002.5 = 145060.5

R2=(−1696.5)2955.9×145060.5=2878112.25138650115.995≈0.020758R^2 = \frac{(-1696.5)^2}{955.9 \times 145060.5} = \frac{2878112.25}{138650115.995} \approx 0.020758

或者,也可以使用 R2=1−SSESYYR^2 = 1 - \frac{SSE}{S_{YY}}。
SSE=∑(Yi−Y^i)2=SYY−β^1SXYSSE = \sum (Y_i - \hat{Y}_i)^2 = S_{YY} - \hat{\beta}_1 S_{XY}
SSE=145060.5−(−1.7747)×(−1696.5)=145060.5−3009.15=142051.35SSE = 145060.5 - (-1.7747) \times (-1696.5) = 145060.5 - 3009.15 = 142051.35
R2=1−142051.35145060.5≈1−0.97924=0.02076R^2 = 1 - \frac{142051.35}{145060.5} \approx 1 - 0.97924 = 0.02076

【答案】R2≈0.0208R^2 \approx 0.0208。

Model 3:複迴歸 Y=α+β1X1+β2X2+β3X3+ϵY = \alpha + \beta_1 X_1 + \beta_2 X_2 + \beta_3 X_3 + \epsilon

(c) 解釋 Model 3 的 F 檢定結果:
F 檢定的虛無假設是所有迴歸係數(除了截距項)同時為零:H0:β1=β2=β3=0H_0: \beta_1 = \beta_2 = \beta_3 = 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 4 題10 分

Consider the hypothesis test that H0:μ≥μ0H_0: \mu \ge \mu_0 v.s. Hα:μ<μ0H_\alpha: \mu < \mu_0, where μ0\mu_0 is a fixed constant. The population variance is known as σ2\sigma^2. Suppose that the significance level is set at α\alpha, and suppose further that the probability of type II error is β\beta when the population mean is μoa\mu_{oa}, where μoa\mu_{oa} is a constant smaller than μ0\mu_0. Define zαz_\alpha and zβz_\beta as numbers satisfying P(Z≥zα)=αP(Z \ge z_\alpha) = \alpha and P(Z≥zβ)=βP(Z \ge z_\beta) = \beta. Derive that the sample size should be (zα+zβμ0−μoa)2σ2\left(\frac{z_\alpha + z_\beta}{ \mu_0 - \mu_{oa}}\right)^2 \sigma^2 (and then round up).

登入後即可作答並保存紀錄。

這一題的完整詳解

此題考驗假設檢定中,第一型錯誤機率、第二型錯誤機率與樣本數的關係,特別是如何推導樣本數的公式。

題目給定的檢定是左尾檢定:
H0:μ≥μ0H_0: \mu \ge \mu_0
Hα:μ<μ0H_\alpha: \mu < \mu_0

顯著水準為 α\alpha,即 P(Reject H0∣μ=μ0)=αP(\text{Reject } H_0 | \mu = \mu_0) = \alpha。
第二型錯誤機率為 β\beta,當實際母體平均數為 μoa\mu_{oa} 時,該機率為 β\beta。 μoa<μ0\mu_{oa} < \mu_0。
β=P(Accept H0∣μ=μoa)=β\beta = P(\text{Accept } H_0 | \mu = \mu_{oa}) = \beta。

樣本數為 nn,母體變異數為 σ2\sigma^2。
檢定統計量為 Z=Xˉ−μ0σ/nZ = \frac{\bar{X} - \mu_0}{\sigma/\sqrt{n}}。

根據顯著水準 α\alpha,我們拒絕 H0H_0 當 Z<−zαZ < -z_\alpha(因為是左尾檢定,臨界值在負方向)。
也就是 Xˉ−μ0σ/n<−zα\frac{\bar{X} - \mu_0}{\sigma/\sqrt{n}} < -z_\alpha,所以拒絕域為 Xˉ<μ0−zασn\bar{X} < \mu_0 - z_\alpha \frac{\sigma}{\sqrt{n}}。

根據第二型錯誤機率 β\beta,當實際母體平均數為 μoa\mu_{oa} 時,我們接受 H0H_0 的機率為 β\beta。
接受域為 Xˉ≥μ0−zασn\bar{X} \ge \mu_0 - z_\alpha \frac{\sigma}{\sqrt{n}}。
所以,β=P(Accept H0∣μ=μoa)=P(Xˉ≥μ0−zασn∣μ=μoa)\beta = P(\text{Accept } H_0 | \mu = \mu_{oa}) = P\left(\bar{X} \ge \mu_0 - z_\alpha \frac{\sigma}{\sqrt{n}} \Big| \mu = \mu_{oa}\right)。

將 Xˉ\bar{X} 轉換為標準常態變數 Z′Z',其平均值為 μoa\mu_{oa},標準誤為 σ/n\sigma/\sqrt{n}。
β=P(Xˉ−μoaσ/n≥(μ0−zασn)−μoaσ/n)\beta = P\left(\frac{\bar{X} - \mu_{oa}}{\sigma/\sqrt{n}} \ge \frac{(\mu_0 - z_\alpha \frac{\sigma}{\sqrt{n}}) - \mu_{oa}}{\sigma/\sqrt{n}}\right)
β=P(Z′≥μ0−μoaσ/n−zα)\beta = P\left(Z' \ge \frac{\mu_0 - \mu_{oa}}{\sigma/\sqrt{n}} - z_\alpha\right)。

題目定義 zβz_\beta 滿足 P(Z≥zβ)=βP(Z \ge z_\beta) = \beta。
然而,這裡的檢定是左尾檢定,接受域是 Xˉ≥c\bar{X} \ge c,拒絕域是 Xˉ<c\bar{X} < c。
在 μ=μoa\mu = \mu_{oa} 時,β=P(Xˉ≥c∣μ=μoa)\beta = P(\bar{X} \ge c | \mu = \mu_{oa})。
標準化後,β=P(Z′≥c−μoaσ/n)\beta = P\left(Z' \ge \frac{c - \mu_{oa}}{\sigma/\sqrt{n}}\right)。

🔒

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23 分

A research study is conducted to compare the kilometers per liter of unleaded regular, mid-grade, and super premium gasolines. There are differences in the performances of different automobiles. Seven different automobiles (numbered in 1 to 7) are sampled randomly and tested. The experimental results, in kilometers per liter, are given in the following table.

🖼️【此處有附圖,請對照原卷】

AutomobileRegularMid-gradeSuper Premium
1262629
2282528
3262830
4292830
5313035
6282933
7283032

(a) To answer the research question, an ANOVA is performed. What are the null and alternative hypotheses for the ANOVA? (2%)
(b) Set up the ANOVA table. (17%)
(c) Using a 0.05 significance level, what is the conclusion? What is your interpretation about the result? (4%)

🖼️ 本題含圖表,以下為原卷對應頁面:
原卷第 3 頁原卷第 4 頁

登入後即可作答並保存紀錄。

這一題的完整詳解

此題考驗單因子變異數分析 (One-Way ANOVA),用於比較三個或更多類別(在這裡是三種汽油類型)的平均值是否存在顯著差異。

(a) 虛無與對立假設:
ANOVA 的目的是檢定各組(汽油類型)的平均值是否相等。

  • H0H_0: 三種汽油類型的平均每公升公里數都相等。
    即 μRegular=μMid-grade=μSuper Premium\mu_{\text{Regular}} = \mu_{\text{Mid-grade}} = \mu_{\text{Super Premium}}。
  • H1H_1: 至少有一種汽油類型的平均每公升公里數與其他不同。
    即 至少存在一對 i,ji, j 使得 μi≠μj\mu_i \ne \mu_j。

【答案】
H0:μRegular=μMid-grade=μSuper PremiumH_0: \mu_{\text{Regular}} = \mu_{\text{Mid-grade}} = \mu_{\text{Super Premium}}
H1:H_1: 至少有一種汽油類型的平均每公升公里數與其他不同。

(b) 建立 ANOVA 表:
ANOVA 表的基本結構包括:變異來源 (Source of Variation)、平方和 (Sum of Squares, SS)、自由度 (Degrees of Freedom, df)、均方 (Mean Square, MS)、F 檢定統計量,以及 P 值。

步驟:

  1. 定義變數:

    • 分組數 (Number of groups, kk) = 3 (Regular, Mid-grade, Super Premium)。
    • 每組樣本數 (Number of observations per group, nin_i) = 7 (有 7 輛汽車)。
    • 總樣本數 (Total number of observations, NN) = k×ni=3×7=21k \times n_i = 3 \times 7 = 21。
  2. 計算各組平均值與總平均值:

    • Regular (組 1):
      • ∑Y1j=26+28+26+29+31+28+28=196\sum Y_{1j} = 26+28+26+29+31+28+28 = 196
      • Yˉ1=196/7=28\bar{Y}_1 = 196 / 7 = 28
    • Mid-grade (組 2):
      • ∑Y2j=26+25+28+28+30+29+30=196\sum Y_{2j} = 26+25+28+28+30+29+30 = 196
      • Yˉ2=196/7=28\bar{Y}_2 = 196 / 7 = 28
    • Super Premium (組 3):
      • ∑Y3j=29+28+30+30+35+33+32=217\sum Y_{3j} = 29+28+30+30+35+33+32 = 217
      • Yˉ3=217/7=31\bar{Y}_3 = 217 / 7 = 31
    • 總平均值 (Yˉ\bar{Y}):
      • ∑Ytotal=196+196+217=609\sum Y_{total} = 196 + 196 + 217 = 609
      • Yˉ=609/21=29\bar{Y} = 609 / 21 = 29
  3. 計算平方和 (Sum of Squares, SS):

    • 組間平方和 (Between Groups SS, SSBSS_B 或 SSTreatmentSS_{Treatment}):
      SSB=∑i=1kni(Yˉi−Yˉ)2SS_B = \sum_{i=1}^{k} n_i (\bar{Y}_i - \bar{Y})^2
      SSB=7×(28−29)2+7×(28−29)2+7×(31−29)2SS_B = 7 \times (28 - 29)^2 + 7 \times (28 - 29)^2 + 7 \times (31 - 29)^2
      SSB=7×(−1)2+7×(−1)2+7×(2)2SS_B = 7 \times (-1)^2 + 7 \times (-1)^2 + 7 \times (2)^2
      SSB=7×1+7×1+7×4=7+7+28=42SS_B = 7 \times 1 + 7 \times 1 + 7 \times 4 = 7 + 7 + 28 = 42

    • 組內平方和 (Within Groups SS, SSWSS_W 或 SSErrorSS_{Error}):
      SSW=∑i=1k∑j=1ni(Yij−Yˉi)2SS_W = \sum_{i=1}^{k} \sum_{j=1}^{n_i} (Y_{ij} - \bar{Y}_i)^2

      • Regular (組 1):
        (26−28)2+(28−28)2+(26−28)2+(29−28)2+(31−28)2+(28−28)2+(28−28)2(26-28)^2 + (28-28)^2 + (26-28)^2 + (29-28)^2 + (31-28)^2 + (28-28)^2 + (28-28)^2
        =(−2)2+02+(−2)2+12+32+02+02=4+0+4+1+9+0+0=18= (-2)^2 + 0^2 + (-2)^2 + 1^2 + 3^2 + 0^2 + 0^2 = 4 + 0 + 4 + 1 + 9 + 0 + 0 = 18
      • Mid-grade (組 2):
        (26−28)2+(25−28)2+(28−28)2+(28−28)2+(30−28)2+(29−28)2+(30−28)2(26-28)^2 + (25-28)^2 + (28-28)^2 + (28-28)^2 + (30-28)^2 + (29-28)^2 + (30-28)^2
        =(−2)2+(−3)2+02+02+22+12+22=4+9+0+0+4+1+4=22= (-2)^2 + (-3)^2 + 0^2 + 0^2 + 2^2 + 1^2 + 2^2 = 4 + 9 + 0 + 0 + 4 + 1 + 4 = 22
      • Super Premium (組 3):
        (29−31)2+(28−31)2+(30−31)2+(30−31)2+(35−31)2+(33−31)2+(32−31)2(29-31)^2 + (28-31)^2 + (30-31)^2 + (30-31)^2 + (35-31)^2 + (33-31)^2 + (32-31)^2
        =(−2)2+(−3)2+(−1)2+(−1)2+42+22+12=4+9+1+1+16+4+1=36= (-2)^2 + (-3)^2 + (-1)^2 + (-1)^2 + 4^2 + 2^2 + 1^2 = 4 + 9 + 1 + 1 + 16 + 4 + 1 = 36
      • SSW=18+22+36=76SS_W = 18 + 22 + 36 = 76
🔒

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