R - 跨数据集的所有行应用函数 (combinevar)

Question

我有一个数据集，其中每一行都包含 combinevar 函数所需的数据（package = fishmethods；combinevar 结合来自两个分布的信息以得出组合方差）。

xbar1 = c(2,2,1,4,3)
xbar2 = c(0,0,0,0,0)
var1 = c(0,1,3,2,1)
var2 = c(0,0,0,0,0)
n1 = c(50,10,30,40,50) 
n2 = c(3,4,50,32,20)

df <- data.frame(xbar1, xbar2, var1, var2, n1, n2)

xbar1 xbar2 var1 var2 n1 n2 
 2     0     0    0   50  3
 2     0     1    0   10  4
 1     0     3    0   30 50
 4     0     2    0   40 32
 3     0     1    0   50 20

我将如何跨行应用函数。我可以像这样在 for 循环中执行此操作：

for (i in 1:nrow(df)) {
    combined_var <- combinevar(xbar = c(df$xbar1[i], df$xbar2[i]), 
       s_squared = c(df$var1[i], df$var2[i]), 
       n = c(df$n1[i], df$n2[i]))[2]
    print(combined_var)
}

[1] 0.2177068
[1] 1.571429
[1] 1.338608
[1] 5.104851
[1] 2.573499

但我确信有更好的方法。我想我可以用一个应用函数来做，但我不知道怎么做。

Answer 1

我们可以按行嵌套数据，然后为每一行映射函数。

library(tidyverse)
library(fishmethods)

df %>% 
  rownames_to_column("row") %>% 
  nest(-row) %>%
  mutate(combined_var = map(data, ~combinevar(xbar = c(.x$xbar1, .x$xbar2), 
                                              s_squared = c(.x$var1, .x$var2), 
                                              n = c(.x$n1, .x$n2))[2])) %>%
  unnest()
#>   row combined_var xbar1 xbar2 var1 var2 n1 n2
#> 1   1    0.2177068     2     0    0    0 50  3
#> 2   2    1.5714286     2     0    1    0 10  4
#> 3   3    1.3386076     1     0    3    0 30 50
#> 4   4    5.1048513     4     0    2    0 40 32
#> 5   5    2.5734990     3     0    1    0 50 20

或者我们可以按行应用函数

df %>% 
  rowwise() %>%
  mutate(combined_var = combinevar(xbar = c(xbar1, xbar2), 
                                              s_squared = c(var1, var2), 
                                              n = c(n1, n2))[2])
#> Source: local data frame [5 x 7]
#> Groups: <by row>
#> 
#> # A tibble: 5 x 7
#>   xbar1 xbar2  var1  var2    n1    n2 combined_var
#>   <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>        <dbl>
#> 1     2     0     0     0    50     3        0.218
#> 2     2     0     1     0    10     4        1.57 
#> 3     1     0     3     0    30    50        1.34 
#> 4     4     0     2     0    40    32        5.10 
#> 5     3     0     1     0    50    20        2.57

由 reprex 创建于 2018-08-19 包 (v0.2.0).

Answer 2

您可以对行使用应用函数并指定函数正确读取行：

    library(fishmethods)

    my_function<- function(vec){
    combined_var <- combinevar(xbar = c(vec[1], vec[2]), s_squared = c(vec[3], vec[4]), n = c(vec[5], vec[6]))
   }

    apply(df, 1, my_function) [2, ]

R - 跨数据集的所有行应用函数 (combinevar)

R - apply a function (combinevar) across all rows of a dataset

r

apply

variance