visualization_distributions.Rmd

---
title: "Data Visualization Distributions"
author: "Md Easin Hasan"
date: "3/27/2021"
output: word_document
---

```{r}

data<-read.csv(file="owid-covid-data.csv",header=TRUE)
dim(data)
#colnames(data)
data1 <- data[data$iso_code == "USA", ] 
data2 <- data[data$iso_code == "GBR", ] 
data3<-rbind(data1,data2)
#dim(data1)
#dim(data2)
country <- ifelse(data3$iso_code == "USA",0,1)
data4 <- cbind(country,data3)
data4$country<-as.factor(data4$country)
data4$country<-factor(data4$country,levels=c(0,1),labels=c("USA", "UK"))
data5 <- data4[,-c(2)]
dat2 <-data5[,c(1,6,8,11,18,20)]
colnames(dat2)
```
```{r}
miss.info<- function(dat, filename=NULL){  
  
  vnames <- colnames(dat); vnames 
  
  n <- nrow(dat)  
  
  out <- NULL  
  
  for (j in 1: ncol(dat)){  
    
    vname <- colnames(dat)[j]  
    
    x <- as.vector(dat[,j]) 
    
    n1 <- sum(is.na(x), na.rm=T)  
    
    n2 <- sum(x=="NA", na.rm=T)  
    
    n3 <- sum(x=="", na.rm=T)  
    
    nmiss <- n1 + n2 + n3  
    
    ncomplete <- n-nmiss 
    
    out <- rbind(out, c(col.number=j, vname=vname, mode=mode(x), n.levels=length(unique(x)), ncomplete=ncomplete, miss.perc=nmiss/n)) } 
  
  out <- as.data.frame(out)  
  
  row.names(out) <- NULL  
  
  return(out) 
  
}  

miss.info(dat2) 
```
```{r}
library(mice)
fit.mice.dat_f <- mice(dat2, m = 1, maxit = 50, method = 'pmm', seed = 500)
dat_f.imputed <- complete(fit.mice.dat_f,1)
datf <- as.data.frame(dat_f.imputed)
```

## Data visualization proportion improvement


## Data visualization proportion improvement



```{r, warning=FALSE}
dat3 <- dat2[,-c(2,3,4,5)]

boxplot(dat3, main = "Covid-19 patients hospitalized in USA and UK",col="orange", border="brown",
        notch = TRUE)
```


## Data visualization distribution improvement

```{r, warning=FALSE}
library(ggplot2)
ggplot(dat2, aes(x = country, y = hosp_patients, fill = country)) +
  geom_boxplot(alpha=0.3) +
    theme(legend.position="right")+
  ggtitle("COVID-19 patients hospitalized in USA & UK")

```