Sort BoxPlot on Date

Question

如何按日期排序？下面

我的數據是：

EXPIRE_DATE是性格，所以我用做發生變異另一真（日期）列。

我覺得我很接近，但是如何按降序或升序排序？

 EXPIRE_DATE  mean  sd   Date 
       (chr) (dbl)  (dbl)   (chr) 
1    04/30/17 56.75132 103.75048  April 2017 
2    08/30/17 30.36706 46.12009 August 2017 
3    08/31/17 42.84366 67.79964 August 2017 
4    12/30/17 26.88593 23.60440 December 2017 
5    12/31/17 38.67540 58.72461 December 2017 
6    02/28/18 42.50570 63.91448 February 2018 
7    01/30/18 28.60205 44.85719 January 2018 
8    01/31/18 70.80121 134.13060 January 2018 
9    07/31/17 45.45389 77.15242  July 2017 
10   06/30/17 47.73592 81.88312  June 2017 
11   05/30/17 46.38233 53.73065  May 2017 
12   05/31/17 52.25520 88.89367  May 2017 
13   11/30/17 39.27158 66.40248 November 2017 
14   10/31/17 40.43197 71.51545 October 2017 
15   09/30/17 43.12762 79.27168 September 2017 

，使代碼如下：

list_mean_sd <- EXPIRING %>% 
       group_by(EXPIRE_DATE) %>% 
       summarize(mean = mean(TOTAL), sd = sd(TOTAL) ) %>% 
       mutate(Date = format(as.Date(EXPIRE_DATE, "%m/%d/%y"), format="%B %Y")) 

我的最終目標是創建一個箱線圖與排序所以它看起來並不怪異的日期..

boxplot(mean ~ Date, data = list_mean_sd, outline = FALSE) 

這是我正在得到的..

dput(head(EXPIRING, 15)) 
structure(list(KEY = c(9495, 9541, 9638, 9717, 9743, 
9921, 10048, 10053, 10061, 10067, 10254, 10343, 24825, 25016, 
25162), TOTAL = c(20, 240, 91.04, 20, 140, 100, 
301.2, 40, 540, 469.82, 40, 140, 133.09, 1700, 20), EXPIRE_DATE = c("11/30/17", 
"01/31/18", "01/31/18", "12/31/17", "12/31/17", "01/31/18", "04/30/17", 
"07/31/17", "01/31/18", "01/31/18", "01/31/18", "01/31/18", "01/31/18", 
"01/31/18", "06/30/17")), .Names = c("KEY", "TOTAL", 
"EXPIRE_DATE"), row.names = c(NA, 15L), class = "data.frame") 

補充說：

dput(head(list_mean_sd, 30)) 
structure(list(EXPIRE_DATE = c("01/30/18", "01/31/18", 
"02/28/18", "04/30/17", "05/30/17", "05/31/17", "06/30/17", "07/31/17", 
"08/30/17", "08/31/17", "09/30/17", "10/31/17", "11/30/17", "12/30/17", 
"12/31/17"), mean = c(28.6020454545455, 70.8012116673021, 42.5057014558283, 
56.751320667367, 46.3823270440252, 52.2552028540308, 47.7359164733179, 
45.4538902012763, 30.3670622064929, 42.843660721111, 43.1276177589063, 
40.4319721861389, 39.2715832825871, 26.8859251197214, 38.6753964550534 
), sd = c(44.857189842357, 134.130597512432, 63.9144788499397, 
103.750483732426, 53.7306532607393, 88.8936749200348, 81.8831227378872, 
77.1524193002944, 46.1200886362958, 67.7996403857795, 79.2716764935199, 
71.5154547562237, 66.4024797158997, 23.6044043594643, 58.7246098554578 
), Date = c("January 2018", "January 2018", "February 2018", 
"April 2017", "May 2017", "May 2017", "June 2017", "July 2017", 
"August 2017", "August 2017", "September 2017", "October 2017", 
"November 2017", "December 2017", "December 2017")), .Names = c("EXPIRE_DATE", 
"mean", "sd", "Date"), class = c("tbl_df", "data.frame"), row.names = c(NA, 
-15L)) 

Answer 1

更新匹配的討論。

您可以強制boxplot按照您希望的順序放置數據，方法是將數據按照您希望的方式排序。

DateOrder = order(as.Date(list_mean_sd$EXPIRE_DATE, "%m/%d/%y")) 
list_mean_sd$Date = factor(list_mean_sd$Date, 
    levels = unique(list_mean_sd$Date[DateOrder])) 
boxplot(mean ~ Date, data = list_mean_sd, cex.axis=0.65)