计算带有两个W变量的X变量的weighted.mean()

Question

我有一个问题，我正在尝试将数据框转换为汇总数据框，其中包含基于多列的总和和加权平均值...

下面是我正在做的一个例子......

df <- data.frame('title' = c("X", "Y", "Z", "X", "Y", "Z"),'date' = c("2020-01-01", "2020-01-01", "2020-01-01", "2020-01-02", "2020-01-03", "2020-01-02"),'weight1' = c(84024, 54241, 106601, 65382, 337007, 687682),'weight2' = c(30, 30, 30, 15, 15, 15),'metric1' = c(3.08, 0.964, 0.839, 1.60, 0.839, 0.648),'metric2' = c(588.03, 298.26, 13.95, 104.29, 10.51, 72.53))

agg = df %>% 
          group_by(df$date, df$title) %>%
          summarise(total_weight1 = sum(weight1, na.rm=TRUE), # total sum
          total_weight1 = sum(weight2, na.rm=TRUE), # total sum
          mean_metric1 = weighted.mean(x = metric1, w = c(weight1, weight2), na.rm=T), # weighted avg
          mean_metric2 = weighted.mean(x = metric2, w = c(weight1, weight2), na.rm=T)) # weighted avg

但是，当然，我收到一条错误消息：

Error in 

总结（）

: 

Problem while computing 

mean_metric1 = 加权.平均值(x = metric1, w = c(权重1, 权重2), na.rm = T)

. The error occurred in group 1: df$date = "2020-01-01", df$title = "X". 

Caused by error in 

weighted.mean.default()

: 

! 'x' and 'w' must have the same length

是否可以计算基于两列的值的加权平均值？我尝试寻找其他来源来手动验证此计算，但失败了......

我也不确定在同一个数据框中计算这一切是否是正确的方法。任何指导将不胜感激。

我尝试过其他来源和函数，包括 rowwise() 和 mutate() 但我不断遇到同样的问题，我的 'x' 和 'w' 需要具有相同的长度......

Answer 1

也许将权重向量相乘会做你除了做的事情

agg = df %>% 
  group_by(df$date, df$title) %>%
  summarise(
    total_weight1 = sum(weight1, na.rm=TRUE), # total sum
    total_weight1 = sum(weight2, na.rm=TRUE), # total sum
    mean_metric1 = weighted.mean(x = metric1, w = weight1 * weight2, na.rm=T), # weighted avg
    mean_metric2 = weighted.mean(x = metric2, w = weight1 * weight2, na.rm=T)) # weighted avg

agg

# A tibble: 6 × 5
# Groups:   df$date [3]
#   `df$date`  `df$title` total_weight1 mean_metric1 mean_metric2
#   <chr>      <chr>              <dbl>        <dbl>        <dbl>
# 1 2020-01-01 X                     30        3.08         588. 
# 2 2020-01-01 Y                     30        0.964        298. 
# 3 2020-01-01 Z                     30        0.839         14.0
# 4 2020-01-02 X                     15        1.6          104. 
# 5 2020-01-02 Z                     15        0.648         72.5
# 6 2020-01-03 Y                     15        0.839         10.5