R 如何获得一辆车的mpg低于其他车的概率？

我用这个来得到汽车mpg的平均值和标准偏差

df1 <- mtcars; df1$rownames = rownames(df1)
df2 <- mtcars; df2$rownames = rownames(df2)
df2$mpg = df2$mpg + rnorm(nrow(df2),0,3)
data = rbind(df1, df2)

---

df1这里比较了
dplyr/data.table
返回概率的方法

library(dplyr)
library(data.table)
library(tidyr)
library(tibble)
# // input data
df <- mtcars[1] %>% 
         rownames_to_column("car")

库（dplyr）
库（数据表）
图书馆（tidyr）
图书馆（tibble）
#//输入数据
df%
行名称到列（“car”）

-测试

# // dplyr
system.time({
        out <- df %>%
                uncount(10000, .id = "run") %>%
                rowwise() %>%
                mutate(sim_mpg = rpois(1, lambda = mpg)) %>%

                group_by(run) %>%
                arrange(sim_mpg) %>%
                mutate(lowest_mpg = row_number() == 1) %>%

                group_by(car) %>%
                summarize(chance_lowest = mean(lowest_mpg),
                orig_mpg = first(mpg))

})
# user system elapsed
# 1.715 0.074 1.787


# // data.table
system.time({

     df_expand <- setDT(df)[rep(seq_len(.N), 10000)][, run := rowid(car)]

     out2 <- df_expand[, sim_mpg := rpois(1, lambda = mpg), 1:nrow(df_expand)
        ][order(sim_mpg), lowest_mpg := seq_len(.N) == 1 ,run
        ][, .(chance_lowest = mean(lowest_mpg), orig_mpg = first(mpg)), .(car)]
})
# user system elapsed
# 0.704 0.050 0.757
sum(out$chance_lowest)
#[1] 1

#//dplyr
系统时间({
超出%
取消计数（10000，.id=“run”）%>%
行（）
变异（sim_mpg=RPOI（1，lambda=mpg））%>%
分组依据（运行）%>%
排列（模拟mpg）%>%
变异（最低值=行数（）==1）%>%
组别(汽车)%>%
总结（机会最低=平均值（最低平均值），
初始值=第一个（mpg））
})
#用户系统运行时间
# 1.715 0.074 1.787
#//data.table
系统时间({
df_扩展
library(dplyr)
library(data.table)
library(tidyr)
library(tibble)
# // input data
df <- mtcars[1] %>% 
         rownames_to_column("car")

# // dplyr
system.time({
        out <- df %>%
                uncount(10000, .id = "run") %>%
                rowwise() %>%
                mutate(sim_mpg = rpois(1, lambda = mpg)) %>%

                group_by(run) %>%
                arrange(sim_mpg) %>%
                mutate(lowest_mpg = row_number() == 1) %>%

                group_by(car) %>%
                summarize(chance_lowest = mean(lowest_mpg),
                orig_mpg = first(mpg))

})
# user system elapsed
# 1.715 0.074 1.787


# // data.table
system.time({

     df_expand <- setDT(df)[rep(seq_len(.N), 10000)][, run := rowid(car)]

     out2 <- df_expand[, sim_mpg := rpois(1, lambda = mpg), 1:nrow(df_expand)
        ][order(sim_mpg), lowest_mpg := seq_len(.N) == 1 ,run
        ][, .(chance_lowest = mean(lowest_mpg), orig_mpg = first(mpg)), .(car)]
})
# user system elapsed
# 0.704 0.050 0.757
sum(out$chance_lowest)
#[1] 1