使用yearmon（）按月份和年份对R中的数据帧进行分组

编辑：

我知道了

df_CloseDelta$YearMonth <- as.yearmon(df_CloseDelta$date)
df_CloseDelta %>%
    group_by(stock, YearMonth) %>%
    summarize(minCloseDelta = min(closeDelta),
              meanCloseDelta = mean(closeDelta),
              maxCloseDelta = max(closeDelta)) -> df_summary_CloseDelta

并返回：

[1] "Jan 2014" "Jan 2014" "Jan 2014" "Jan 2014" "Jan 2014" "Jan 2014" "Jan 2014"
[8] "Jan 2014" "Jan 2014" "Jan 2014" "Jan 2014" "Jan 2014" "Jan 2014" "Jan 2014"
[15] "Jan 2014" "Jan 2014" "Jan 2014" "Jan 2014" "Jan 2014" "Jan 2014" "Jan 2014"
[22] "Jan 2014" "Jan 2014" "Jan 2014" "Jan 2014" "Jan 2014" "Jan 2014" "Jan 2014"

等等

然后我尝试将其分组：

df_summary_CloseDelta <- df_CloseDelta %>%
    group_by(as.yearmon(df_CloseDelta$date))

---

我知道有1006个日期，但有5030个条目，因为有五只股票。我试着对它们进行分组，然后找出每个股票每月和每年的平均值、最小值和最大值。有人能给我指出正确的方向吗？

group\u by

希望您为其指定变量名，或与数据中的行数相同的向量，该行数将被视为执行分组的因子。请参见下面的示例

> btest <- data.frame(a = LETTERS[1:10],
+                     b = c(1,1,2,2,3,3,4,4,5,5),
+                     c = c(rep('e',5), rep('f',5)))
> btest
   a b c
1  A 1 e
2  B 1 e
3  C 2 e
4  D 2 e
5  E 3 e
6  F 3 f
7  G 4 f
8  H 4 f
9  I 5 f
10 J 5 f

然而，您的代码认为您正在尝试做的是提供它将用于形成分组的逐行值

> btest %>% 
+   group_by(c(1,1,1,1,1,2,2,2,2,2)) %>% 
+   summarise(ex = mean(b))
# A tibble: 2 x 2
  `c(1, 1, 1, 1, 1, 2, 2, 2, 2, 2)`    ex
                              <dbl> <dbl>
1                              1.00  1.80
2                              2.00  4.20

这里的问题是，您需要先添加要按其分组的列，然后才能按其分组

> df_CloseDelta[['date_yearmon']] <- as.yearmon(df_CloseDelta[['date']])
> 
> df_CloseDelta %>% 
+   group_by(date_yearmon, stock) %>% 
+   summarise(mean_closedelta = mean(closeDelta))
# A tibble: 240 x 3
# Groups:   date_yearmon [?]
   date_yearmon  stock mean_closedelta
   <S3: yearmon> <chr>           <dbl>
 1 Jan 2014      AAPL          -0.474 
 2 Jan 2014      AMZN          -0.472 
 3 Jan 2014      FB             0.746 
 4 Jan 2014      GOOG           0.310 
 5 Jan 2014      MSFT           0.104 
 6 Feb 2014      AAPL           0.269 
 7 Feb 2014      AMZN           0.0631
 8 Feb 2014      FB             0.491 
 9 Feb 2014      GOOG           0.159 
10 Feb 2014      MSFT           0.0713
# ... with 230 more rows

---

xts有

to.monthly

，它直接转换为monthly，因此假设输入的OHLCV数据位于环境

e

中的一组xts对象中，如注释所示，最后我们对

e

中的每个此类对象应用一个转换函数（将两者转换为monthly、转换为数据帧并附加符号）然后对得到的数据帧进行rbinding，得到一个数据帧

sym2df <- function(x, env) cbind(Symbol = x, fortify.zoo(to.monthly(env[[x]], name = "")))
do.call("rbind", lapply(ls(e), sym2df, env = e))

> btest %>% 
+   group_by(c) %>% 
+   summarise(ex = mean(b))
# A tibble: 2 x 2
  c        ex
  <fct> <dbl>
1 e      1.80
2 f      4.20

> btest %>% 
+   group_by(c(1,1,1,1,1,2,2,2,2,2)) %>% 
+   summarise(ex = mean(b))
# A tibble: 2 x 2
  `c(1, 1, 1, 1, 1, 2, 2, 2, 2, 2)`    ex
                              <dbl> <dbl>
1                              1.00  1.80
2                              2.00  4.20

> mean(c(1,1,2,2,3))
[1] 1.8
> mean(c(3,4,4,5,5))
[1] 4.2

> df_CloseDelta[['date_yearmon']] <- as.yearmon(df_CloseDelta[['date']])
> 
> df_CloseDelta %>% 
+   group_by(date_yearmon, stock) %>% 
+   summarise(mean_closedelta = mean(closeDelta))
# A tibble: 240 x 3
# Groups:   date_yearmon [?]
   date_yearmon  stock mean_closedelta
   <S3: yearmon> <chr>           <dbl>
 1 Jan 2014      AAPL          -0.474 
 2 Jan 2014      AMZN          -0.472 
 3 Jan 2014      FB             0.746 
 4 Jan 2014      GOOG           0.310 
 5 Jan 2014      MSFT           0.104 
 6 Feb 2014      AAPL           0.269 
 7 Feb 2014      AMZN           0.0631
 8 Feb 2014      FB             0.491 
 9 Feb 2014      GOOG           0.159 
10 Feb 2014      MSFT           0.0713
# ... with 230 more rows

df_CloseDelta %>%
  mutate(date_yearmon = as.character(as.yearmon(date))) %>%
  group_by(date_yearmon, stock) %>%
  summarise(mean_closedelta = mean(closeDelta))

sym2df <- function(x, env) cbind(Symbol = x, fortify.zoo(to.monthly(env[[x]], name = "")))
do.call("rbind", lapply(ls(e), sym2df, env = e))

library(quantmod)

start <- "2014-01-01"
end <- "2017-12-31"
syms <- c("AAPL", "AMZN", "FB", "GOOG", "MSFT")
getSymbols(syms, from = start, to = end, env = e <- new.env())