How to calculate 7 day moving average ?Alternative way

Question

I replicated a code to plot 7-day moving average of currently hospitalized people in every state.

This is my code:

covus %>%
  filter(measure == "hospitalized_currently", state %in% unique(state_pops$state)) %>%
  group_by(state) %>%
  mutate(hospitalized7 = slider::slide_dbl(count, mean, .before = 7, .after = 0, na.rm = TRUE)) %>%
  left_join(state_pops) %>%
  ggplot(mapping = aes(x = date, y = (hospitalized7/pop)*1e5)) + 
  geom_line(size = 0.5) + 
  scale_y_continuous(labels = scales::comma_format(accuracy = 1)) + 
  facet_wrap(~ name, ncol = 4) +
  labs(x = "Date", 
       y = "Hospitalized per 100,000 Population (Seven Day Rolling Average)")

And this is the outcome plot

As you can see any curve is readable.

I want something like this:

So I thought that 7-day moving average is not in my plot. Is there a way to know that? In addition, Can you give me any suggestion to make my curves readable?

Make your plot taller, or plot fewer states as in the example. You need to make more room to actually draw the graphs. A ggplot will try to fill whatever space you give it so give it more space. — MrFlick, Jun 26 '20 at 22:06
If you provide a sample of your data or a minimum reproducible example ([How to make a great R reproducible example](https://stackoverflow.com/questions/5963269/how-to-make-a-great-r-reproducible-example)) you will have a far better chance of getting help with this. — jared_mamrot, Jun 26 '20 at 23:50

score 0 · Accepted Answer · answered Jun 27 '20 at 00:43

A very similar approach of what you want can be done using data at this link(I did a similar plot few days ago so you can adapt your dataset): https://www.sendspace.com/file/4ikz5d

The approach uses dplyr and zoo. I hope it can help. Just download data from previous link.

library(ggplot2)
library(dplyr)
library(reshape2)
library(tidyr)
library(stringr)
library(ggcharts)
library(zoo)
library(xlsx)
library(scales)
library(ggrepel)
#Load data
df4 <- read.csv('df4.csv')
df4$Day <- as.Date(df4$Day)
#Create rolling variable
df4 %>% group_by(country) %>% mutate(NewCases=c(0,diff(confirmed)),
                                     RollAvg=rollmean(NewCases,k = 7,fill = NA,align = 'right')) -> df4
#Plot
ggplot(df4, aes(x = Day ,y = RollAvg)) +geom_line(size=1.2,color='blue')+
  scale_x_date(date_breaks = "25 day",date_labels = "%b/%d")+
  theme_bw()+xlab("Date")+ylab('New confirmed cases (Movil mean of 7 days)')+
  facet_wrap(.~country,scales = 'free')+
  theme(axis.text.x = element_text(face="bold"),
        axis.text.y = element_text(face="bold"),
        plot.title = element_text(hjust = 0.5,size=16,face="bold"),
        plot.subtitle = element_text(size=12,face="bold"),
        strip.text.x = element_text(size = 10, face = "bold"),
        legend.position = "none")

How to calculate 7 day moving average ?Alternative way

1 Answers1