[英]How to calculate the difference between mean and the value of a variable for all columns with dyplr
我想计算数据集中所有列的平均值和值之间的差异。 我的等式是:
residual = y - mean(y)
我的原始数据集如下所示:
d_season_combined %>% head()
Player Pos Age Team G GS MPG FG FGA FG% 3P 3PA 3P% 2P 2PA 2P% eFG% FT FTA FT% ORB DRB TRB
1 Álex Abrines SG 24 OKC 75 8 15.1 1.5 3.9 0.395 1.1 2.9 0.380 0.4 0.9 0.443 0.540 0.5 0.6 0.848 0.3 1.2 1.5
2 Quincy Acy PF 27 BRK 70 8 19.4 1.9 5.2 0.356 1.5 4.2 0.349 0.4 1.0 0.384 0.496 0.7 0.9 0.817 0.6 3.1 3.7
3 Steven Adams C 24 OKC 76 76 32.7 5.9 9.4 0.629 0.0 0.0 0.000 5.9 9.3 0.631 0.629 2.1 3.8 0.559 5.1 4.0 9.0
4 Bam Adebayo C 20 MIA 69 19 19.8 2.5 4.9 0.512 0.0 0.1 0.000 2.5 4.8 0.523 0.512 1.9 2.6 0.721 1.7 3.8 5.5
5 Arron Afflalo SG 32 ORL 53 3 12.9 1.2 3.1 0.401 0.5 1.3 0.386 0.7 1.7 0.413 0.485 0.4 0.5 0.846 0.1 1.2 1.2
6 Cole Aldrich C 29 MIN 21 0 2.3 0.2 0.7 0.333 0.0 0.0 NA 0.2 0.7 0.333 0.333 0.1 0.3 0.333 0.1 0.6 0.7
我已经像这样计算了每一列的平均值:
d_league_average <- d_season_combined %>%
select(-c(Player, Team, Pos, Season)) %>%
mutate_if(is.character, as.numeric) %>%
summarise_all(mean, na.rm = TRUE)
Age G GS MPG FG FGA FG% 3P 3PA 3P% 2P 2PA
1 25.99439 48.79159 22.99065 19.27589 3.113551 6.873738 0.4432531 0.8437383 2.40972 0.3118051 2.269533 4.465794
我已经设法像这样计算单个列的平均残差:
d_season_combined %>%
select(PTS) %>%
lapply(function(i)i-d_league_average$PTS)
$PTS
[1] -3.68504673 -2.48504673 5.51495327 -1.48504673 -4.98504673 -7.78504673 14.71495327 -0.18504673 -7.28504673
[10] -3.68504673 0.91495327 -2.18504673 -0.48504673 0.91495327 -7.18504673 18.51495327 7.81495327 -2.48504673
但我不知道如何为所有列做到这一点。 我试过用 summarise_all 解决这个问题:
d_residuals <- d_season_combined %>%
select(-c( Pos, Team)) %>%
group_by(Season, Player) %>%
summarise_all(function(i)d_league_average[i])
带 lapply:
d_residuals <- lapply(d_season_combined[column_names], function(i) y-d_league_average[i]) %>% bind_rows()
但这对我不起作用。 我怎样才能为所有列实现这一目标? 感谢您的帮助
与其单独计算d_league_average
然后使用它来计算d_residuals
,我们可以直接在一个步骤中从其各自的列均值中减去这些值。
library(dplyr)
d_season_combined %>%
select(-c(Player, Team, Pos)) %>%
mutate_if(is.character, as.numeric) %>%
mutate_all(~. - mean(., na.rm = TRUE))
声明:本站的技术帖子网页,遵循CC BY-SA 4.0协议,如果您需要转载,请注明本站网址或者原文地址。任何问题请咨询:yoyou2525@163.com.