110 92 104 110 方差齐性检验 2 两组方差齐性的检验命令(仅适合两组方差齐性检验) sdtest x, by(group) Variance ratio test Group Me Std. Err. Std. Dev. [95% Conf. Intervall 0 89.081.8229289.1146485.3176692.84234 101.521.9009829.50491197.59657105.4434 combined 95.31.5774561.154392.1299898.47002 F(24, 24)observed F ob =0.920 F(24, 24) lower tail =FL F ob 0.920 F(24, 24)upper tail FU =1/F obs= 1.087 Ha: sd(0)< sd (1) Ha (0)=sd(1) Ha: sd(0)>sd(1) P<Fobs=0.4195P<FL+P>FU=0.8389P>Fobs=0.5805 P值=0.8389α,因此可以认为两组方差齐性的
1 98 1 110 1 89 1 103 1 89 1 121 1 94 1 95 1 92 1 109 1 98 1 98 1 120 1 104 1 110 方差齐性检验 H0：σ1＝σ2 vs H1:σ1≠σ2 α=0.1 两组方差齐性的检验命令(仅适合两组方差齐性检验) sdtest x,by(group) Variance ratio test ------------------------------------------------------------------------------ Group | Obs Mean Std. Err. Std. Dev. [95% Conf. Interval] ---------+-------------------------------------------------------------------- 0 | 25 89.08 1.822928 9.11464 85.31766 92.84234 1 | 25 101.52 1.900982 9.504911 97.59657 105.4434 ---------+-------------------------------------------------------------------- combined | 50 95.3 1.577456 11.1543 92.12998 98.47002 ------------------------------------------------------------------------------ Ho: sd(0) = sd(1) F(24,24) observed = F_obs = 0.920 F(24,24) lower tail = F_L = F_obs = 0.920 F(24,24) upper tail = F_U = 1/F_obs = 1.087 Ha: sd(0) < sd(1) Ha: sd(0) ~= sd(1) Ha: sd(0) > sd(1) P < F_obs = 0.4195 P < F_L + P > F_U = 0.8389 P > F_obs = 0.5805 P 值=0.8389>>α，因此可以认为两组方差齐性的
正态性检验:H:资料服从正态分布ⅴsH:资料偏态分布 每一组资料正态性检验 ktest x if group==0 Skewness/Kurtosis tests for Normality Jo1 Variable Pr(Skewness) Pr(Kurtosis) adj chi2(2) Prob>chi2 0.927 0.326 1.0 0.5926 sktest x if group==l s/Kurtosis tests for Normality Variable Pr(Skewness) Pr(Kurtosis) adj chi2(2) Prob>chi2 0.474 0.675 0.73 0.6948 P值均大于α,因此可以认为两组资料都服从正态分布 H:p1=μ2vsH1:μu1≠2 =0.05 ttest x, by(group) Two-sample t test with equal variances oup Obs Mean Std. Err. Std. Dev. [95% Conf. Interval] 89.081.8229289.1146485.3176692.84234 25101.521.9009829.50491197.59657105.4434 combined 1.57745611.154392.1299898.47002 diff I -12.442.633781 17.73557-7.144429 Degrees of freedom: 48 Ho: mean (0)- mean (1)=diff =0 Ha: diff< 0 Ha: diff=0 Ha: diff>0 4.7232 t=-4.7232 P<t=0.0000 P>|t=0.0000 =1.0000
正态性检验：H0：资料服从正态分布 vs H1：资料偏态分布 α=0.05 每一组资料正态性检验 sktest x if group==0 Skewness/Kurtosis tests for Normality ------- joint ------ Variable | Pr(Skewness) Pr(Kurtosis) adj chi2(2) Prob>chi2 -------------+------------------------------------------------------- x | 0.927 0.326 1.05 0.5926 . sktest x if group==1 Skewness/Kurtosis tests for Normality ------- joint ------ Variable | Pr(Skewness) Pr(Kurtosis) adj chi2(2) Prob>chi2 -------------+------------------------------------------------------- x | 0.474 0.675 0.73 0.6948 P 值均大于α，因此可以认为两组资料都服从正态分布 H0：μ1＝μ2 vs H1：μ1≠μ2 α=0.05 ttest x,by(group) Two-sample t test with equal variances ------------------------------------------------------------------------------ Group | Obs Mean Std. Err. Std. Dev. [95% Conf. Interval] ---------+-------------------------------------------------------------------- 0 | 25 89.08 1.822928 9.11464 85.31766 92.84234 1 | 25 101.52 1.900982 9.504911 97.59657 105.4434 ---------+-------------------------------------------------------------------- combined | 50 95.3 1.577456 11.1543 92.12998 98.47002 ---------+-------------------------------------------------------------------- diff | -12.44 2.633781 -17.73557 -7.144429 ------------------------------------------------------------------------------ Degrees of freedom: 48 Ho: mean(0) - mean(1) = diff = 0 Ha: diff < 0 Ha: diff ~= 0 Ha: diff > 0 t = -4.7232 t = -4.7232 t = -4.7232 P < t = 0.0000 P > |t| = 0.0000 P > t = 1.0000