我有一个包含6列的数据框(X):mean.x,s.x,n.x,mean.y,s.y,n.y。它们代表来自群体x和y的平均值,st dev(s)和样本大小(n)。我正在运行一个R包(BSDA),它根据这些统计参数执行t检验。问题是,对于每一行,我得到1个汇总表,我有640.000行。
我想要做的是创建包含640.000汇总表中所有p值和其他参数的新列。这可能吗?
前5行的值相同:mean.x(0.444357),sx(0.02575427),nx(633744),mean.y(0.4308),sy(0.000628747),ny(390)
这是显示摘要表的脚本:
library(BSDA)
tsum.test(mean.x = X$mean.x,
s.x = X$s.x,
n.x = X$n.x,
mean.y = X$mean.y,
s.y = X$s.y,
n.y = X$n.y,
alternative = "less",
mu = 0, # null hypothesis that there is no diff between means
var.equal = FALSE,
conf.level = 0.95)
非常感谢!
答案 0 :(得分:1)
可能是的。看看下面。一种方法是使用apply
:
想象一下这样一个非常简单的data.frame
(对于这个简单的例子,所有行都是相同的):
x <- c(7.8, 6.6, 6.5, 7.4, 7.3, 7.0, 6.4, 7.1, 6.7, 7.6, 6.8)
y <- c(4.5, 5.4, 6.1, 6.1, 5.4, 5.0, 4.1, 5.5)
X <- data.frame(mean_x = mean(x), s.x = sd(x), n.x = 11, mean_y = mean(y), s.y = sd(y),
n.y = 8)
X <- rbind(X, X, X)
#> X
# mean_x s.x n.x mean_y s.y n.y
#1 7.018182 0.4643666 11 5.2625 0.7069805 8
#2 7.018182 0.4643666 11 5.2625 0.7069805 8
#3 7.018182 0.4643666 11 5.2625 0.7069805 8
然后,您使用apply
在每一行上运行tsum.test
,然后提取所需的参数。对于示例,我提取了p.values
和degrees of freedom
:
new_cols <-
apply(X, 1, function(x) {
#using apply in each iteration, a row will be fed to the tsum.test function
#so make sure you re using the correct ones
stats <-
#x[1] corresponds to the first column, x[2] to the second and so on
tsum.test(mean.x = x[1],
s.x = x[2],
n.x = x[3],
mean.y = x[4],
s.y = x[5],
n.y = x[6],
alternative = "less",
mu = 0, # null hypothesis that there is no diff between means
var.equal = FALSE,
conf.level = 0.95)
#output p.values and degrees of freedom on this occasion
c(pvalue = stats$p.value, df = stats$parameters)
})
上面输出了自由度和p.values,为了绑定到你的data.frame,你可以做到:
> cbind(X, t(new_cols))
mean_x s.x n.x mean_y s.y n.y pvalue.mean_x df.df
1 7.018182 0.4643666 11 5.2625 0.7069805 8 0.9999669 11.30292
2 7.018182 0.4643666 11 5.2625 0.7069805 8 0.9999669 11.30292
3 7.018182 0.4643666 11 5.2625 0.7069805 8 0.9999669 11.30292