IntroToRWithCovidEDA.Rmd

---
title: "Intro to R and COVID-19 EDA"
author: "Julia Layne"
date: "6/29/2020"
output: html_document
---

# Resources for Learning R


### [DataCamp Intro To R](https://www.datacamp.com/courses/free-introduction-to-r)
  - Truly introductory
  - Code In Browser, has hints

### [Install R](https://cran.revolutionanalytics.com)
  - I grabbed the Texas Mirror, but any will work on this page
  - [R Install Mirrors](https://cran.r-project.org/mirrors.html)

### [RStudio](https://rstudio.com)

### [RStudio Cloud](https://rstudio.cloud/)
  - Allows you to use R before installing. You can work on projects from any computer with a browser.
  - [Out of Beta August 2020 - Plans and Pricing](https://rstudio.cloud/plans/free)

### [FreeCodeCamp Intro Video](https://www.youtube.com/watch?v=_V8eKsto3Ug)
  - Made by Barton Poulson
  - [His R Intro](https://datalab.cc/tools/r01)
  - [Intro to R Playlist](https://www.youtube.com/watch?list=PLkk92zzyru5OOYKXfC4OWzc4Lzo_lBOLP&time_continue=22&v=Cl-9aLU292Y&feature=emb_title), I suggest starting at the plot. You can go back to the previous 



# R Basics

This is an R Markdown file. It lets you run by code block or output a report with knittr

## Math

```{r}
print("1 + 1")
1 + 1
print("13* 13")
13* 13
print("25 / 5")
25 / 5
print("21 / 5")
21 / 5
print("21 %% 5")
21 %% 5
print("5^3")
5^3
```

## Variables

```{r}
5 -> five
thirteen = 13
print("1 + 1")
1 + 1
print("13* 13")
thirteen* thirteen
print("25 / 5")
(20 + five) / five
print("21 / 5")
21 / five
print("21 %% 5")
21 %% five
print("5^3")
five^3
```

## Vectors

```{r}
names <- c('test','Myocarditis','beta coronavirus','novel corona 2019','Encephalitis','Hepatitis A', 'influenza', 'coronaitis')
names

which_name <- function(name) {
  type <- 'Neither'
  if(grepl('corona', name,ignore.case = TRUE)){
    type <-'corona'
    }
  if(grepl('itis', name,ignore.case = TRUE)){
    type <-'itis'
    }
  type 
}

#testing top dataset to confirm it is working
sapply(names, which_name,simplify = TRUE)
```


# COVID DATASET

>  - From the [John Hopkins dataset](https://github.com/CSSEGISandData/COVID-19) avaiable on GitHub
>  - COVID-19/csse_covid_19_data/csse_covid_19_daily_reports/


## Read in Files
Change with new COVID file for most up to date info
```{r}
covid_raw = read.csv("06-28-2020-us.csv",header = TRUE)
#covid_ts_confirmed = read.csv(file.choose(),header = TRUE)
```

## Look at Raw Data

```{r}
head(covid_raw)
#covid_ts_confirmed
```

# Plot Data

Initial lookinto ploting data points
Note here
```{r}
#install.packages("tidyverse")
library(tidyverse)
library(ggplot2)
library(GGally)
library(plotly)

plot(covid_raw)

byState10 <- covid_raw %>% head(10) %>% ggplot(aes(x=Deaths, y=Confirmed, fill=Province_State)) + geom_point(aes(fill=Province_State)) + ggtitle("Deaths vs Confirmed Cases in First Ten Provinces")

ggplotly(byState10)

byState <- covid_raw %>% ggplot(aes(x=Deaths, y=Confirmed, fill=Province_State)) + geom_point(aes(fill=Province_State)) + ggtitle("Deaths vs Confirmed Cases per Province")
ggplotly(byState)

byStateNoNY <- filter(covid_raw, Province_State != 'New York' & Province_State != 'New Jersey')  %>% ggplot(aes(x=Deaths, y=Confirmed, fill=Province_State)) + geom_point(aes(fill=Province_State)) + ggtitle("Deaths vs Confirmed Cases per Province")
ggplotly(byStateNoNY)

#covid_raw %>% head(10) %>% ggplot(aes(x=Deaths, y=Confirmed, fill=Province_State) + geom_point(aes(fill=Province_State)) + ggtitle("Deaths vs Confirmed Cases in Top Countries (Minus China)")

```

```{r}
statesAbove700Deaths <- filter(covid_raw, Deaths >= 700)  %>% ggplot(aes(x=Deaths, y=Confirmed, fill=Province_State)) + geom_point(aes(fill=Province_State)) + ggtitle("Over 700 Mortality States")
ggplotly(statesAbove700Deaths)

statesAbove700Deaths <- filter(covid_raw, Deaths < 700)  %>% ggplot(aes(x=Deaths, y=Confirmed, fill=Province_State)) + geom_point(aes(fill=Province_State)) + ggtitle("Under 700 Mortality States")
ggplotly(statesAbove700Deaths)
```



# Map data using Fips codes

```{r}
library(e1071)
library(usmap)

UScovid_dataset <- filter(covid_raw, Country_Region == 'US' & FIPS != 'NA') 
UScovid_dataset

#UScovid_dataset$fips <- fips(brew_count_by_state$state)
attach(UScovid_dataset)
UScovid_dataset_fips <- UScovid_dataset[order(FIPS),] 
detach(UScovid_dataset)


UScovid_dataset_fips$fips = UScovid_dataset_fips$FIPS

plot_usmap(data = UScovid_dataset_fips,  
           values = "Deaths", 
           color = rgb(.2, .7, 1)) + 
    labs(title = "Covid Deaths by State", 
         subtitle = "Count of Covid19 Deaths per state") + 
  scale_fill_continuous(low = "white", high = rgb(.2, .7, 1), 
                        name = "Deaths per state", label = scales::comma) + theme(legend.position = "right")

plot_usmap(data = filter(UScovid_dataset_fips, Province_State != 'New York'),  values = "Deaths", color = rgb(.2, .7, 1)) + 
    labs(title = "Covid Deaths by State (New York Removed)", subtitle = "Count of Covid19 Deaths per state") + 
  scale_fill_continuous(low = "white", high = rgb(.2, .7, 1), name = "Deaths per state", label = scales::comma) + theme(legend.position = "right")

plot_usmap(data = filter(UScovid_dataset_fips, Province_State != 'New York' & Province_State != 'New Jersey'),  values = "Deaths", color = rgb(.2, .7, 1)) + 
    labs(title = "Covid Deaths by State (New York Removed)", subtitle = "Count of Covid19 Deaths per state") + 
  scale_fill_continuous(low = "white", high = rgb(.2, .7, 1), name = "Deaths per state", label = scales::comma) + theme(legend.position = "right")
```

# Previous Work - Finding the top Countries by Confirmed Cases

## Read in Files
Change with new COVID file for most up to date info
```{r}
covid_raw_world = read.csv("06-28-2020.csv",header = TRUE)
#covid_ts_confirmed = read.csv(file.choose(),header = TRUE)
head(covid_raw_world)
```


```{r}
confirmed_by_country <- covid_raw_world%>% group_by(Country_Region) %>% tally(Confirmed, name = "Confirmed", sort = TRUE)
head(confirmed_by_country)

deaths_by_country <- covid_raw_world%>% group_by(Country_Region) %>% tally(Deaths, name = "Deaths", sort = TRUE)
head(deaths_by_country)

totals <- merge(confirmed_by_country, deaths_by_country, by="Country_Region")
head(totals)

```

Then reordered by Confirmed
```{r}

#order(totals$Confirmed, decreasing = TRUE)
#totals$Confirmed
#totals[180,]

top_to_least <- totals[order(totals$Confirmed, decreasing = TRUE),]

head(top_to_least)
```

```{r}

top10Confirmed <- top_to_least %>% head(10) %>% ggplot(aes(x=Deaths, y=Confirmed, fill=Country_Region)) + geom_point(aes(fill=Country_Region)) + ggtitle("Deaths vs Confirmed Cases in Top countries")

ggplotly(top10Confirmed)

# At the time, China was the highest and I wanted to look at the rest, now it is much different
top10ConfirmedMinusUS <- subset(top_to_least, Country_Region != "US")  %>% head(10) %>% ggplot(aes(x=Deaths, y=Confirmed, fill=Country_Region)) + geom_point(aes(fill=Country_Region)) + ggtitle("Deaths vs Confirmed Cases in Top Countries (Minus US)")
ggplotly(top10ConfirmedMinusUS)

# Now removing US instead
top10ConfirmedMinusUSandB <- subset(top_to_least, Country_Region != "US" & Country_Region != "Brazil")  %>% head(10) %>% ggplot(aes(x=Deaths, y=Confirmed, fill=Country_Region)) + geom_point(aes(fill=Country_Region)) + ggtitle("Deaths vs Confirmed Cases in Top Countries (Minus US and Brazil)")
ggplotly(top10ConfirmedMinusUSandB)
```