4、
临界值法
在显著水平 α = 0.05 \alpha=0.05 α=0.05下检验假设 H 0 : μ ≤ 10 , H 1 : μ > 10 H_0:\mu\le 10,H_1:\mu>10 H0:μ≤10,H1:μ>10
n = 20 , X ˉ = 10.2 , S = 0.5099 n=20,\bar X=10.2,S=0.5099 n=20,Xˉ=10.2,S=0.5099
t 0.05 ( n − 1 ) = t 0.05 ( 19 ) = 1.7291 t_{0.05}(n-1)=t_{0.05}(19)=1.7291 t0.05(n−1)=t0.05(19)=1.7291
拒绝域为 t = X ˉ − μ 0 S / n ≥ t α ( n − 1 ) = 1.7291 t=\frac{\bar X-\mu_0}{S/\sqrt{n}}\geq t_{\alpha}(n-1)=1.7291 t=S/nXˉ−μ0≥tα(n−1)=1.7291
t t t的观察值为 10.2 − 10 0.5099 / 2 = 1.754 > 1.7291 \frac{10.2-10}{0.5099/\sqrt{2}}=1.754>1.7291 0.5099/210.2−10=1.754>1.7291
即:落在拒绝域内,在 α = 0.05 \alpha=0.05 α=0.05下拒绝 H 0 H_0 H0
P值法
P { t ≥ t 0 } = P { t ≥ 1.754 } = 0.04776 < 0.05 P\{t\ge t_0\}=P\{t\ge 1.754\}=0.04776<0.05 P{t≥t0}=P{t≥1.754}=0.04776<0.05
x=c(9.8,10.4,10.6,9.6,9.7,9.9,10.9,11.1,9.6,10.2,10.3,9.6,9.9,11.2,10.6,9.8,10.5,10.1,10.5,9.7)
m=mean(x)
# 临界值法
alpha=0.05
(m-10)/sd(x)*sqrt(20) > qt(alpha,19,lower.tail = FALSE)
# p值法
1-pt((m-10)/sd(x)*sqrt(20),19)
7、
临界值法
检验假设 H 0 : μ 1 − μ 2 ≤ 2 , H 1 : μ 1 − μ 2 > 2 H_0:\mu_1-\mu_2\le 2,H_1:\mu_1-\mu_2>2 H0:μ1−μ2≤2,H1:μ1−μ2>2
X ∼ N ( μ 1 , σ 2 ) , Y ∼ N ( μ 2 , σ 2 ) X\sim N(\mu_1,\sigma^2),Y\sim N(\mu_2,\sigma^2) X∼N(μ1,σ2),Y∼N(μ2,σ2)
X ˉ = 5.25 , Y ˉ = 1.5 , S 1 2 = 0.9318 , S 2 2 = 1 , S W = 0.9828 \bar X=5.25,\bar Y=1.5,S_1^2=0.9318,S_2^2=1,S_W=0.9828 Xˉ=5.25,Yˉ=1.5,S12=0.9318,S22=1,SW=0.9828
t t t检验, α = 0.05 \alpha=0.05 α=0.05
拒绝域为 W = { X ˉ − Y ˉ S w 1 n 1 + 1 n 2 > t α ( n 1 + n 2 − 2 } W=\{\frac{\bar X-\bar Y}{S_w\sqrt{\frac{1}{n_1}+\frac{1}{n_2}}}>t_\alpha(n_1+n_2-2\} W={Swn11+n21Xˉ−Yˉ>tα(n1+n2−2}
t 0 . 05 ( 22 ) = 1.7171 t_0.05(22)=1.7171 t0.05(22)=1.7171
样本值为 5.25 − 1.5 − 2 0.99828 1 12 + 1 12 = 4.3616 > 1.7171 \frac{5.25-1.5-2}{0.99828\sqrt{\frac{1}{12}+\frac{1}{12}}}=4.3616>1.7171 0.99828121+1215.25−1.5−2=4.3616>1.7171
拒绝原假设
p值法
P { t ≥ t 0 } = P { t ≥ 4.3616 } = 0.000125 < 0.05 P\{t\ge t_0\}=P\{t\ge 4.3616\}=0.000125<0.05 P{t≥t0}=P{t≥4.3616}=0.000125<0.05
拒绝原假设
x=c(6,4,5,5,6,5,5,6,4,6,7,4)
y=c(2,1,2,2,1,0,3,2,1,0,1,3)
# 临界值法
alpha=0.05
(mean(x)-mean(y)-2)/0.9828/sqrt(1/12+1/12) > qt(alpha,19,lower.tail = FALSE)
# p值法
1-pt(4.3616,19)
17、
(1)
H 0 : σ 1 2 = σ 2 2 , H 1 : σ 1 2 ≠ σ 2 2 H_0:\sigma_1^2=\sigma_2^2,H_1:\sigma_1^2\neq\sigma_2^2 H0:σ12=σ22,H1:σ12=σ22
S 1 2 S 2 2 = 1.09 \frac{S_1^2}{S_2^2}=1.09 S22S12=1.09
F 0.025 ( 9 , 9 ) = 4.03 , F 0.975 ( 9 , 9 ) = 0.248 F_{0.025}(9,9)=4.03,F_{0.975}(9,9)=0.248 F0.025(9,9)=4.03,F0.975(9,9)=0.248
0.248 < 1.9 < 4.03 0.248<1.9<4.03 0.248<1.9<4.03 接受原假设
拒绝域为 S 1 2 S 2 2 ≥ F 0.025 ( 99 ) ∣ ∣ S 1 2 S 2 2 ≤ F 0.975 ( 9.9 ) \frac{S_1^2}{S_2^2}\ge F_{0.025}(99) || \frac{S_1^2}{S_2^2}\le F_{0.975}(9.9) S22S12≥F0.025(99)∣∣S22S12≤F0.975(9.9)
P P P值为 P 0 = 0.55 > 0.05 P_0=0.55>0.05 P0=0.55>0.05
接受原假设
x=c(101,100,99,99,98,100,98,99,99,99)
y=c(100,98,100,99,98,99,98,98,99,100)
# 临界值法
v1=var(x)/var(y)
v1 >= qf(0.025,9,9) && v1 = qf(1-0.025,9,9)
# p值法
pf(v1,9,9)
(2)
σ 1 2 = σ 2 2 = σ 2 σ \sigma_1^2=\sigma_2^2=\sigma^2\quad\sigma σ12=σ22=σ2σ未知
X ˉ = 99.2 , Y ˉ = 98.9 , S 1 2 = 0.805 , S 2 2 = 0.77 , S w 2 = 0.805 , ∣ t ∣ = 0.748 \bar X=99.2,\bar Y=98.9, S_1^2=0.805,S_2^2=0.77,S_w^2=0.805,|t|=0.748 Xˉ=99.2,Yˉ=98.9,S12=0.805,S22=0.77,Sw2=0.805,∣t∣=0.748
拒绝域为 ∣ t ∣ ≥ t α 2 ( n 1 + n 2 − 2 ) = t 0.025 ( 18 ) = 2.1009 > 0.748 |t|\ge t_{\frac{\alpha}{2}(n_1+n_2-2)}=t_{0.025}(18)=2.1009>0.748 ∣t∣≥t2α(n1+n2−2)=t0.025(18)=2.1009>0.748
接受原假设
P 0 = P { ∣ t ∣ ≥ t 0 } = 0.4645 < 0.05 P_0=P\{|t|\ge t_0\}=0.4645<0.05 P0=P{∣t∣≥t0}=0.4645<0.05
接受原假设
x=c(101,100,99,99,98,100,98,99,99,99)
y=c(100,98,100,99,98,99,98,98,99,100)
# 临界值法
alpha=0.05
(mean(x)-mean(y))/((9*var(x)+9*var(y))/18)/sqrt(1/9+1/9) > qt(alpha/2,18,lower.tail = FALSE)
# p值法
1-pt(0.748,18)