如何在同一画布中绘制每月和每日数据

Question

I have a dataset with over 5,000 records here https://login.filesanywhere.com/fs/v.aspx?v=8c6b678858616ea970a0

My last column (sal) is monthly data and my other columns are daily data. I have tried many ways to plot it together in the same r device but my 'sal' column prints some weird lines. My other graphics are plotted correctly. Could some one give me a hint on how I can summarize 'sal' to be only one value per month to plot it on my bottom graphic using facet_wrap? Here is some code that I have tried:

 library(tidyverse)
 library(readxl)
 library(lubridate)
tab <- read_excel("myjoinedfile.xlsx", sheet="Sheet1")
test <- tab %>% pivot_longer(-c("year","day","date","month")) %>% data.frame()
ggplot(test, aes(x = date, y = value,colour=year)) + 
  geom_line(size=0.75) +  t
  facet_wrap(name~.,switch = "y",ncol=1) +  
  scale_x_date(breaks = seq(as.Date("1991-10-01"),as.Date("1992-09-30"),
    by = "1 month"),date_labels = "%b") 

All my graphics print correctly but the bottom one is not displaying correctly because is so many repeated values on column 'sal'. I unsuccessfully tried to get the mean with pivot_longer() Thanks beforehand.

Answer 1

This could be achieved like so:

1. Aggregate you column sal by month and year , add the name and the date . Here I set the date to the first of the month. But you could change that.
2. Convert your data to long except column sal
3. Bind the two df together.

library(tidyverse)
library(readxl)
library(lubridate)
tab <- read_excel("myjoinedfile.xlsx", sheet="Sheet1")

test1 <- select(tab, year, month, sal) %>%
  group_by(year, month) %>% 
  summarise(value = mean(sal), .groups = "drop") %>% 
  mutate(name = "sal", 
         date = ifelse(month < 10, 1992, 1991),
         date = paste(date, month, "01", sep = "-"),
         date = as.Date(date))
  
test2 <- tab %>% 
  select(-sal) %>% 
  pivot_longer(-c("year","day","date","month")) %>% 
  mutate(date = as.Date(date))

test <- bind_rows(test1, test2)

ggplot(test, aes(x = date, y = value,colour=year)) + 
  geom_line(size=0.75) +
facet_wrap(name~.,switch = "y",ncol=1) +  
  scale_x_date(breaks = seq(as.Date("1991-10-01"),as.Date("1992-09-30"),
                            by = "1 month"),date_labels = "%b")