This analysis shows the impact of Events(i.e Lightning, Tornado, TSTM Wind etc)

R-packages used in the below analysis - ggplot2,dplyr

Synopsis

The Storm events causes various damages across United States of America and its consequences/damanges have been recorded across various healt and property areas. The analysed data shows that Tornado has the top health as well as economic impact on people and lives by a large margin.

Data Processing

The data collected is from various NOAA Storm Database and stored as a csv file which is then read. The EVTYPE column or characteristic is converted into “character type” to perform further action accordingly. 2 new columns are added- TOTAL_AFFECTED,TOTAL_ECONOMICDMG which combines FATALITIES,INJURIES and PROPDMG,CROPDMG respectively. The data is then grouped on EVTYPE and Top 5 most harmful EVTYPE are recorded and plotted for Comparision.

1. Downloading the data

library("ggplot2")
library("dplyr")

## 
## Attaching package: 'dplyr'

## The following objects are masked from 'package:stats':
## 
##     filter, lag

## The following objects are masked from 'package:base':
## 
##     intersect, setdiff, setequal, union

download.file("https://d396qusza40orc.cloudfront.net/repdata%2Fdata%2FStormData.csv.bz2","Storm.csv.bz2")

2. Reading the data

data<-read.csv("Storm.csv.bz2")

3. Adding new columns(aggregated) for analysis

  data<-mutate(data,TOTAL_AFFECTED=data$FATALITIES+data$INJURIES)
  data<-mutate(data,TOTAL_ECONOMICDMG=data$PROPDMG+data$CROPDMG)
  data$EVTYPE<-as.character(data$EVTYPE)
  data$EVTYPE<-toupper(data$EVTYPE)

4. Grouping the data and choosing Top5 Damage creating events

  grouped_by_events<-data %>% group_by(EVTYPE) %>% summarise(TOTAL.AFFECTED=sum(TOTAL_AFFECTED,na.rm = TRUE),TOTAL.ECONOMICDMG=sum(TOTAL_ECONOMICDMG,na.rm = TRUE))
  top5<-tail(order(grouped_by_events$TOTAL.AFFECTED),5)
  ##selecting only top 5 data
  grouped_by_events<-grouped_by_events[top5,]
  grouped_by_events$EVTYPE<-as.factor(grouped_by_events$EVTYPE)

5. Plotting the data

par(mfrow=c(2,1))
  ##plotting 1st plot
  barplot(names.arg=grouped_by_events$EVTYPE,height = grouped_by_events$TOTAL.AFFECTED,xlab="Event Type",ylab="Total Affected",col=grouped_by_events$EVTYPE,main="Health Damange Chart",space=5)
  ##plotting 2nd plot
  barplot(names.arg=grouped_by_events$EVTYPE,height = grouped_by_events$TOTAL.ECONOMICDMG,xlab="Event Type",ylab="Economic Damage",col=grouped_by_events$EVTYPE,main="Economic Damange Chart",space=5)

6. Complete R-Code

##downloading the data from URL
  download.file("https://d396qusza40orc.cloudfront.net/repdata%2Fdata%2FStormData.csv.bz2","Storm.csv.bz2")
  ##reading the downloaded data
  data<-read.csv("Storm.csv.bz2")
  ##add new columns
  data<-mutate(data,TOTAL_AFFECTED=data$FATALITIES+data$INJURIES)
  data<-mutate(data,TOTAL_ECONOMICDMG=data$PROPDMG+data$CROPDMG)
  data$EVTYPE<-as.character(data$EVTYPE)
  data$EVTYPE<-toupper(data$EVTYPE)
  ##grouping the data
  grouped_by_events<-data %>% group_by(EVTYPE) %>% summarise(TOTAL.AFFECTED=sum(TOTAL_AFFECTED,na.rm = TRUE),TOTAL.ECONOMICDMG=sum(TOTAL_ECONOMICDMG,na.rm = TRUE))
  top5<-tail(order(grouped_by_events$TOTAL.AFFECTED),5)
  ##selecting only top 5 data
  grouped_by_events<-grouped_by_events[top5,]
  grouped_by_events$EVTYPE<-as.factor(grouped_by_events$EVTYPE)
  par(mfrow=c(2,1))
  ##plotting 1st plot
  barplot(names.arg=grouped_by_events$EVTYPE,height = grouped_by_events$TOTAL.AFFECTED,xlab="Event Type",ylab="Total Affected",col=grouped_by_events$EVTYPE,main="Health Damange Chart",space=5)
  ##plotting 2nd plot
  barplot(names.arg=grouped_by_events$EVTYPE,height = grouped_by_events$TOTAL.ECONOMICDMG,xlab="Event Type",ylab="Economic Damage",col=grouped_by_events$EVTYPE,main="Economic Damange Chart",space=5)