Load libraries and Data

library(covid19.analytics)
library(data.table)
library(sqldf)
library(plotly)
library(openair)
## https://cran.r-project.org/web/packages/covid19.analytics/vignettes/covid19.analytics.html
##https://www.kaggle.com/modesty520/covid-19-detailed-visualization

iuh_colors <- c("#990000", "#EEEDEB","#01426A")
covid_data <- covid19.US.data()

## ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 
## -------------------------------------------------------------------------------- 
## ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 
## --------------------------------------------------------------------------------

indiana <- subset(covid_data, Province_State == 'Indiana')
arkansas <- subset(covid_data, Province_State == 'Arkansas')

setDT(indiana)


indiana_long <- melt(indiana, id.var = c("Country_Region", "Province_State",
                                           "Lat", "Long_"))


indiana_summary <- sqldf("select variable as date, sum(value) as cases
                           from indiana_long
                           group by variable")

setDT(indiana_summary)
indiana_summary[ , new := cases - shift(cases)]


in_plot <- plot_ly(indiana_summary, x = ~date, y = ~cases, 
                   type = 'scatter', 
                   mode = 'lines',
                   colors=iuh_colors,
                   hoverinfo='text',
                   text = ~paste('</br> Date: ', date,
                                 '</br> Cases: ' , prettyNum(cases, big.mark=",",scientific=FALSE))
)

in_plot_new <- plot_ly(indiana_summary, x = ~date, y = ~new, 
                   type = 'bar', 
                   colors=iuh_colors,
                   hoverinfo='text',
                   text = ~paste('</br> Date: ', date,
                     '</br> Cases: ' , prettyNum(new, big.mark=",",scientific=FALSE),
                     '</br> Cumulative: ', prettyNum(cases, big.mark=",",scientific=FALSE))
)

Cumulative Cases:

This is helpful to get a quick overview of cases over time. You can really see the growth taking off in October.

in_plot <- plot_ly(indiana_summary, x = ~date, y = ~cases, 
                   type = 'scatter', 
                   mode = 'lines',
                   fill='tozeroy',
                   hoverinfo='text',
                   line = list(color=iuh_colors[1]),
                   fillcolor = list(color=iuh_colors[1]),
                   text = ~paste('</br> Date: ', date,
                                 '</br> Cases: ' , prettyNum(cases, big.mark=",",scientific=FALSE))
) %>% layout(title = 'Cumulative Covid Cases (Indiana)', xaxis=list(autotick=F, dtick=30))

in_plot

New Cases

In this plot, you can also see the growth. In addition, you can see as the new cases start to reach the peak and slowly start decreasing.

in_plot_new <- plot_ly(indiana_summary, x = ~date, y = ~new, 
                   type = 'bar', 
                   marker = list(color=iuh_colors[1]),
                   hoverinfo='text',
                   text = ~paste('</br> Date: ', date,
                                 '</br> Cases: ' , prettyNum(new, big.mark=",",scientific=FALSE))
) %>% layout(title = 'New Covid Cases (Indiana)', xaxis=list(autotick=F, dtick=30))

in_plot_new

New Cases - Calendar View

I find this view very intuitive. This really helps highlight how much higher the case counts were in the fall than the early peaks we saw in spring/summer.

date <- seq(from=as.Date(min(indiana_summary$date)),
            by=1,
            to = as.Date(max(indiana_summary$date)))


in_new <- indiana_summary[,"new"]
cal_plot <- calendarPlot(data.frame(in_new, date), pollutant = 'new', year = 2020, main = "New Cases")