从R中的数据框获取净值作为比例 [英] Getting net values as a proportion from a dataframe in R

问题描述

我在R（p2.df）中有一个数据框，该数据框已将一系列值聚合为以下内容（还有更多列，这只是一个简写版本）：

I have a dataframe in R (p2.df) that has aggregated a range of values into the following (there are many more columns this is just an abridge version):

genre       rating  cc      dd      ee
Adventure   FAILURE 140393  20865   358806
Adventure   SUCCESS 197182  32872   492874
Fiction     FAILURE 140043  14833   308602
Fiction     SUCCESS 197725  28848   469879
Sci-fi      FAILURE 8681    1682    24259
Sci-fi      SUCCESS 7439    1647    22661

我想获取每列比例的净值，我可以在电子表格中获得它，但在R studio中却不能。

I want to get the net values of the proportions for each column, which I can get in a spreadsheet but can't in R studio.

电子表格中的公式遵循以下模式：

The formula in the spreadsheet follows the pattern:

net_cc = (cc(success)/(cc(success)+dd(success)+ee(success)) - (cc(fail)/(cc(fail)+dd(fail)+ee(fail))

我想在R中得到的是可以从电子表格中获得的这张表：

What I want to get out in R is this table that I can get from the spreadsheet:

genre       net_cc          net_dd          net_ee
Adventure   0.002801373059  0.005350579467  -0.008151952526
Fiction     -0.01825346696  0.009417699223  0.008835767735
Sci-fi      -0.01641517271  0.003297091109  0.0131180816

有什么想法吗？如果有什么用，我通过将上一张表总结为p2.df来创建它：

Any ideas how? If it's any use I created the p2.df by summarising a previous table as:

library(dplyr)

p2.df<- s2.df %>% group_by(genre,rating) %>% summarise_all(sum)  

推荐答案

使用 tidyverse ：

library(tidyverse)
df %>% gather(,,3:5) %>%
  spread(rating,value) %>%
  group_by(genre) %>%
  transmute(key,net = SUCCESS/sum(SUCCESS) - FAILURE/sum(FAILURE)) %>%
  ungroup %>%
  spread(key,net)

# # A tibble: 3 x 4
# genre           cc      dd       ee
#   <chr>        <dbl>   <dbl>    <dbl>
# 1 Adventure  0.00280 0.00535 -0.00815
# 2 Fiction   -0.0183  0.00942  0.00884
# 3 Sci-fi    -0.0164  0.00330  0.0131 

这篇关于从R中的数据框获取净值作为比例的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！