country_minning.Rmd

---
date: "4/29/2020"
output: html_document
---

```{r setup, include=FALSE}
knitr::opts_chunk$set(fig.align = "center", fig.width = 8, fig.height=10, echo = FALSE, message=FALSE, warning=FALSE)
```

```{r run code}
library(readr)
library(tidyverse)
library(readxl)
library(zoo)
library(ggthemes)
library(extrafont)
library(kableExtra)
library(lubridate)


my_theme <- theme_clean()

loadfonts(device = "win")

number_of_countries <- 35
number_of_countries2 <- 35
number_of_cols <- 4
tspan <- 120

# excluded_countries <- c("Qatar", "Ecuador")
excluded_countries <- c()

#load data 

#load population for countries around the world. 

pop_lookup <- read_rds(here::here("pop_country_level.rds")) %>% 
  mutate(Country_Region = ifelse(Country_Region == "Korea, South", "South Korea", Country_Region))

df1 <- read_csv("data_world_download/COVID-19 Activity.csv",
col_types = cols(REPORT_DATE = col_date(format = "%Y-%m-%d"),
COUNTY_NAME = col_character(), COUNTY_FIPS_NUMBER = col_character(),
PEOPLE_POSITIVE_CASES_COUNT = col_number(),
PEOPLE_POSITIVE_NEW_CASES_COUNT = col_number(),
PEOPLE_DEATH_COUNT = col_number(),
PEOPLE_DEATH_NEW_COUNT = col_number()))   %>% 
  select(Date = REPORT_DATE,
         Admin2 = COUNTY_NAME,
         Province_State = PROVINCE_STATE_NAME,
         Country_Region = COUNTRY_SHORT_NAME,
         country_code = COUNTRY_ALPHA_2_CODE,
         CONTINENT_NAME,
         Cases = PEOPLE_POSITIVE_CASES_COUNT,
         Difference = PEOPLE_POSITIVE_NEW_CASES_COUNT,
         Deaths = PEOPLE_DEATH_NEW_COUNT,
         Cumulative_Deaths = PEOPLE_DEATH_COUNT,
         ) %>%  
  group_by(Date, Country_Region) %>%  
  summarize_if(is.numeric, sum) %>%  
  left_join(pop_lookup, by = "Country_Region")

n_reginos <- length(unique(df1$Country_Region))


#find factor by cases
order1 <- df1 %>% 
  ungroup() %>% 
  group_by(Country_Region) %>% 
  filter(Date == max(Date)) %>%  
  select(Country_Region, Cases) %>%  
  arrange(desc(Cases)) 

ln <- length(order1$Country_Region)

order1 <- order1 %>%   
ungroup() %>% 
mutate(orderv = 1:ln) %>% 
  select(-Cases)



#========================================================

df1 <- df1 %>% 
    left_join(order1, by = "Country_Region") %>% 
  ungroup() %>% 
  mutate(Country_Region = factor(Country_Region)) %>%  
  mutate(Country_Region = fct_reorder(Country_Region, orderv)) %>% 
  filter(orderv <= 60) 

top_countries <- df1 %>%  
  ungroup() %>%
  group_by(Country_Region) %>%  
  filter(Date == max(Date)) %>%  
  select(Country_Region, Cases) %>% 
  ungroup() %>%  
  top_n(number_of_countries)

df1 <- df1 %>% 
  filter(Country_Region %in% top_countries$Country_Region)

#daily cases
df1 %>%  
  ggplot(aes(x = Date, y = Difference, colour = Country_Region)) +  
  facet_wrap(~ Country_Region, ncol = number_of_cols) +
  geom_smooth(se = FALSE) +
  geom_segment(aes(x = Date, y = Difference , xend = Date, yend = 0), alpha = 0.3) + 
  geom_point(aes(x = Date, y = Difference), alpha = 0.4) +
  my_theme +
  theme(legend.position = "none",
        axis.text.x = element_text(angle = 45, hjust = 1)) +
  labs(title = "Daily Cases")

#cumulative cases
df1 %>% 
  ggplot(aes(x = Date, y = Cases, colour = Country_Region)) +  
  facet_wrap(~ Country_Region, ncol = number_of_cols) +
  geom_smooth(se = FALSE) +
  geom_segment(aes(x = Date, y = Cases , xend = Date, yend = 0), alpha = 0.3) + 
  geom_point(aes(x = Date, y = Cases), alpha = 0.4) +
  my_theme +
  theme(legend.position = "none",
        axis.text.x = element_text(angle = 45, hjust = 1)) +
  labs(title = "Cumulative Cases")

#daily cases, y is free
df1 %>% 
  ggplot(aes(x = Date, y = Difference, colour = Country_Region)) +  
  facet_wrap(~ Country_Region, ncol = number_of_cols, scales = "free") +
  geom_smooth(se = FALSE) +
  geom_segment(aes(x = Date, y = Difference , xend = Date, yend = 0), alpha = 0.3) + 
  geom_point(aes(x = Date, y = Difference), alpha = 0.4) +
  my_theme +
  theme(legend.position = "none",
        axis.text.x = element_text(angle = 45, hjust = 1)) +
  labs(title = "Daily Cases", subtitle = "Y-axsis is free")


#cases per capita 


order2 <- df1 %>% 
  mutate(Difference_pc = (Difference/population)*100000) %>% 
  ungroup() %>%  
  group_by(Country_Region) %>% 
  top_n(7, Date) %>% 
  arrange(Country_Region, desc(Difference_pc)) %>%  
  select(Date, Country_Region, Difference_pc) %>% 
  group_by(Country_Region) %>%  
  summarize(Difference_pc = mean(Difference_pc)) %>% 
  arrange(desc(Difference_pc)) %>%  
  select(Country_Region)


l2 <- length(order2$Country_Region)

order2 <- order2 %>% 
  mutate(orderlvl2 = 1:l2)
  
df1 %>% 
  mutate(Difference_pc = (Difference/population)*100000) %>% 
  filter(!(Country_Region %in% excluded_countries)) %>% 
  left_join(order2, by = "Country_Region") %>% 
  mutate(Country_Region = fct_reorder(Country_Region, orderlvl2)) %>% 
  ggplot(aes(x = Date, y = Difference_pc, colour = Country_Region)) +  
  facet_wrap(~ Country_Region, ncol = number_of_cols, scales = "fixed") +
  geom_smooth(se = FALSE) +
  geom_segment(aes(x = Date, y = Difference_pc , xend = Date, yend = 0), alpha = 0.3) + 
  geom_point(aes(x = Date, y = Difference_pc), alpha = 0.4) +
  ylim(0,75) +
  my_theme +
  theme(legend.position = "none",
        axis.text.x = element_text(angle = 45, hjust = 1)) +
  labs(title = "Daily Cases per 100,000 people", 
       subtitle = "Arranged by most weekly-average cases per captia ")

df1 %>% 
  mutate(Difference_pc = (Difference/population)*100000) %>% 
  ggplot(aes(x = Date, y = Difference_pc, colour = Country_Region)) +  
  facet_wrap(~ Country_Region, ncol = number_of_cols, scales = "free") +
  geom_smooth(se = FALSE) +
  geom_segment(aes(x = Date, y = Difference_pc , xend = Date, yend = 0), alpha = 0.3) + 
  geom_point(aes(x = Date, y = Difference_pc), alpha = 0.4) +
  my_theme +
  theme(legend.position = "none",
        axis.text.x = element_text(angle = 45, hjust = 1)) +
  labs(title = "Daily Cases per 100,000 people", subtitle = "Y-axsis is free")


df1 %>% 
  filter(Deaths >= 0) %>% 
  mutate(Deaths_pc = (Deaths/population)*100000) %>% 
  ggplot(aes(x = Date, y = Deaths_pc, colour = Country_Region)) +  
  facet_wrap(~ Country_Region, ncol = number_of_cols, scales = "fixed") +
  geom_smooth(se = FALSE) +
  geom_segment(aes(x = Date, y = Deaths_pc , xend = Date, yend = 0), alpha = 0.3) + 
  geom_point(aes(x = Date, y = Deaths_pc), alpha = 0.4) +
  my_theme +
  theme(legend.position = "none",
        axis.text.x = element_text(angle = 45, hjust = 1)) +
  labs(title = "Daily Fatalities per 100,000 people")

df1 %>% 
  filter(Cumulative_Deaths >= 0) %>% 
  mutate(Cumulative_Deaths_pc = (Cumulative_Deaths/population)*100000) %>% 
  ggplot(aes(x = Date, y = Cumulative_Deaths_pc, colour = Country_Region)) +  
  facet_wrap(~ Country_Region, ncol = number_of_cols, scales = "fixed") +
  geom_smooth(se = FALSE) +
  geom_segment(aes(x = Date, y = Cumulative_Deaths_pc , xend = Date, yend = 0), alpha = 0.3) + 
  geom_point(aes(x = Date, y = Cumulative_Deaths_pc), alpha = 0.4) +
  my_theme +
  theme(legend.position = "none",
        axis.text.x = element_text(angle = 45, hjust = 1)) +
  labs(title = "Cumulative Fatalities per 100,000 people")

```