生物统计13.pptVIP

生物统计学 2009年度 计算生物所 非参数检验 6月19日 例10.4 Wilcoxon signed-rank test wilcox.test(A,B) Wilcoxon rank sum test with continuity correction data: A and B W = 54, p-value = 0.02364 alternative hypothesis: true location shift is not equal to 0 例10.7 weight=c(55, 49, 42, 21, 52, 61, 112, 30, 89, 63, 42, 97, 81, 95, 92, 169, 137, 168, 85, 154) group=as.factor(c(rep(1:4,each=5))) Kruskal-Wallis rank sum test kruskal.test(weight~group) Kruskal-Wallis rank sum test data: weight by group Kruskal-Wallis chi-squared = 11.4114, df = 3, p-value = 0.009697 例10.8 beat=c(85, 81, 75, 74, 85, 83, 86, 77, 76, 89, 80, 83, 76, 75, 88, 82, 78, 78, 80, 82) type=as.factor(c(rep(1:4,each=5))) subject=as.factor(c(rep(1:5,4))) Friedman rank sum test friedman.test(beat~type|subject) Friedman rank sum test data: beat and type and subject Friedman chi-squared = 3.48, df = 3, p-value = 0.3234 Pearson定理 趋向于自由度为r－1的卡方分布。 Goodness of Fit. 例10.10 chisq.test(c(335, 125, 160),p=c(9,3,4)/16) Chi-squared test for given probabilities data: c(335, 125, 160) X-squared = 1.362, df = 2, p-value = 0.5061 例10.11 prop.test(35,35+46,p=0.5,correct=F) 1-sample proportions test without continuity correction data: 35 out of 35 + 46, null probability 0.5 X-squared = 1.4938, df = 1, p-value = 0.2216 alternative hypothesis: true p is not equal to 0.5 95 percent confidence interval: 0.3297208 0.5406256 sample estimates: p 0.4320988 例10.11 prop.test(35,35+46, p=0.5) 1-sample proportions test with continuity correction data: 35 out of 35 + 46, null probability 0.5 X-squared = 1.2346, df = 1, p-value = 0.2665 alternative hypothesis: true p is not equal to 0.5 95 percent confidence interval: 0.3240072 0.5466829 sample estimates: p 0.4320988 列联表的检验 卡方检验 Fisher严格检验 例10.15 M=matrix(c(252,224,145,136,103,140),c(2,3)) M [,1] [,2] [,3] [1,] 252 145 103 [2,] 224 136 140 例10.15 chisq.test(M) Pearsons Chi-squared test data: M X-squared = 7.5691, df = 2, p-value =