covid19_trajectory_plot.R

#---------------------------------------------------------
# Creating a trajectory plot for covid19 confirmed cases
# Required packages - dplyr, tidyr, and plotly packages
# Data - coronavirus package
#---------------------------------------------------------

# To get the most update data use the Github version
#devtools::install_github("RamiKrispin/coronavirus")
library(coronavirus)
`%>%` <- magrittr::`%>%`

# Trajectory function
covid_trajec <- function(min_case = 100, max_days = 80, country){
  country_df <- lapply(country, function(i){
    df <- coronavirus %>% dplyr::filter(type == "confirmed", Country.Region == i) %>%
      dplyr::group_by(date) %>%
      dplyr::summarise(cases = sum(cases)) %>%
      dplyr::ungroup() %>%
      dplyr::arrange(date) %>%
      dplyr::mutate(total = cumsum(cases),
                    country = i) %>%
      dplyr::filter(total > min_case)  %>%
      dplyr::select(-cases, -date)
    
    if(nrow(df) > max_days){
      df <- df[1:max_days,]
    }
    
    df$index <- 1:nrow(df)
    return(df)
  }) %>% dplyr::bind_rows() %>% 
    tidyr::pivot_wider(names_from = country, values_from = total) %>%
    as.data.frame()
  
  p <- plotly::plot_ly()
  
  for(i in 2:ncol(country_df)){
    p <- p %>% 
      plotly::add_lines(x = country_df$index,
                        y = country_df[, i],
                        name = names(country_df)[i],  line = list(width = 2))
  }
  
  # Set or modify here the layout/titles
  p <- p %>%  plotly::layout(title = "Covid19 Confirmed Cases Trajectory Plot",
  yaxis = list(title = "Cumulative Positive Cases",type = "log"),
  xaxis = list(title = paste("Days since the total positive cases surpass", min_case)),
  hovermode = "compare")
  
  return(p)

} 


# Example for function setting
min_case <- 100
max_days <- 80
country <- c("China", "Italy", "Iran", "United Kingdom", "Spain", "France", "Russia", "Korea, South", "US")

covid_trajec(min_case = min_case, max_days = max_days, country = country)

This is very nice.