Synopsis:

The following analysis is has two goals. 1. Find across the United States, which types of events are most harmful with respsect to health. 2. Find across the United States, which types of events have the greatest economic consequences.

Required Packages:

library(knitr)

Now firstly we need to load the data:

stormData <- read.csv("repdata-data-StormData.csv")

Data Processing:

Determine the top ten causes of fatalaties and injuries across America.

usFatalitiesMean <- with(stormData, tapply(FATALITIES,EVTYPE,mean))
usFatalitiesTopTenPerEvent <- sort(usFatalitiesMean, decreasing = TRUE)[1:10]

usInjuriesMean <- with(stormData, tapply(INJURIES,EVTYPE,mean))
usInjuriesTopTenPerEvent <- sort(usInjuriesMean, decreasing = TRUE)[1:10]

usFatalitiesTotal <- with(stormData, tapply(FATALITIES,EVTYPE,sum))
usFatalitiesTopTenTotal <- sort(usFatalitiesTotal, decreasing = TRUE)[1:10]

usInjuriesTotal <- with(stormData, tapply(INJURIES,EVTYPE,sum))
usInjuriesTopTenTotal <- sort(usInjuriesTotal, decreasing = TRUE)[1:10]

Now to find the greatest single event that caused the most fatalities/injuries.

greatestFatality <- stormData[which.max(stormData$FATALITIES),]
greatestFatalityDate <- as.character(greatestFatality$BGN_DATE)


greatestInjuries <- stormData[which.max(stormData$INJURIES),]
greatestInjuriesDate <- as.character(greatestInjuries$BGN_DATE)

Analysis from above claims that Tornados do the most damage to health. Let’s see what states they occur in.

usTornadoesByState <- with(stormData[stormData$EVTYPE == "TORNADO",], 
                           tapply(FATALITIES,STATE,sum))
usTornadoesByStateTopTen <- sort(usTornadoesByState, decreasing = TRUE)[1:10]

For Economic Evaluation need to convert costs of each event to like terms.

moneyConverter <- function(xDMG,xEXP){
        if (xEXP == "k" || xEXP == "K"){
                xDMG = xDMG*1e3
        } else if (xEXP == "m" || xEXP == "M"){
                xDMG = xDMG*1e6
        } else if (xEXP == "B") {
                xDMG = xDMG*1e9
        } else if (xEXP == "h" || xEXP == "H") {
                xDMG = xDMG*1e2
        }
        
        if(is.numeric(xEXP)) {
                xDMG = xDMG^xEXP
        }
        
        return (xDMG)
}

expensesByEvent <- with(stormData, tapply(
        moneyConverter(PROPDMG,PROPDMGEXP)
        + moneyConverter(CROPDMG,CROPDMGEXP),EVTYPE,sum))

## Warning in if (xEXP == "B") {: the condition has length > 1 and only the
## first element will be used

expensesByEventTopTen <- sort(expensesByEvent, decreasing = TRUE)[1:10]
Million <- 1e6
expensesByEventTopTenInBillions <- expensesByEventTopTen/Million

Results:

kable(as.list(usFatalitiesTopTenPerEvent), col.names = names(usFatalitiesTopTenPerEvent), 
      format = "html", digits = 1, caption = "Events with the Highest Average 
      Number of Fatalities per event", align = 'c')

Events with the Highest Average Number of Fatalities per event
TORNADOES, TSTM WIND, HAIL	COLD AND SNOW	TROPICAL STORM GORDON	RECORD/EXCESSIVE HEAT	EXTREME HEAT	HEAT WAVE DROUGHT	HIGH WIND/SEAS	MARINE MISHAP	WINTER STORMS	Heavy surf and wind
25	14	8	5.7	4.4	4	4	3.5	3.3	3

kable(as.list(usInjuriesTopTenPerEvent), 
      col.names = names(usInjuriesTopTenPerEvent),
      format = "html", digits = 1, caption = "Ten Highest Averages of US Injuries
      per Event", align = 'c')

Ten Highest Averages of US Injuries per Event
Heat Wave	TROPICAL STORM GORDON	WILD FIRES	THUNDERSTORMW	HIGH WIND AND SEAS	SNOW/HIGH WINDS	GLAZE/ICE STORM	HEAT WAVE DROUGHT	WINTER STORM HIGH WINDS	HURRICANE/TYPHOON
70	43	37.5	27	20	18	15	15	15	14.5

kable(as.list(usFatalitiesTopTenTotal), 
      col.names = names(usInjuriesTopTenTotal),
      format = "html", digits = 1, caption = "Events with the Highest Total Number 
      of Fatalities", align = 'c')

Events with the Highest Total Number of Fatalities
TORNADO	TSTM WIND	FLOOD	EXCESSIVE HEAT	LIGHTNING	HEAT	ICE STORM	FLASH FLOOD	THUNDERSTORM WIND	HAIL
5633	1903	978	937	816	504	470	368	248	224

kable(as.list(usInjuriesTopTenTotal), col.names = names(usInjuriesTopTenTotal),
      format = "html", digits = 1, caption = "Events with the Highest Total Number
      of Injuries", align = 'c')

Events with the Highest Total Number of Injuries
TORNADO	TSTM WIND	FLOOD	EXCESSIVE HEAT	LIGHTNING	HEAT	ICE STORM	FLASH FLOOD	THUNDERSTORM WIND	HAIL
91346	6957	6789	6525	5230	2100	1975	1777	1488	1361

sprintf("The Greatest Fatality occured in %s, which was a/an %s on %s
        ", greatestFatality$STATE, greatestFatality$EVTYPE
        ,strsplit(greatestFatalityDate," ")[[1]][1])

## [1] "The Greatest Fatality occured in IL, which was a/an HEAT on 7/12/1995\n        "

sprintf("The Greatest Injuries occured in %s, which was a/an %s on %s
        ", greatestInjuries$STATE, greatestInjuries$EVTYPE
        ,strsplit(greatestInjuriesDate," ")[[1]][1])

## [1] "The Greatest Injuries occured in TX, which was a/an TORNADO on 4/10/1979\n        "

barplot(usTornadoesByStateTopTen, main = "Total number of deaths by Tornadoes per State",
        ylab = "Number of deaths", xlab = "State")

Economic Results:

kable(as.list(expensesByEventTopTenInBillions), 
      col.names = names(expensesByEventTopTenInBillions),
      format = "html", digits = 2, caption = "Events with Highest Expenses (in millions)",
      align = 'c')

Events with Highest Expenses (in millions)
TORNADO	FLASH FLOOD	TSTM WIND	FLOOD	THUNDERSTORM WIND	HAIL	LIGHTNING	THUNDERSTORM WINDS	HIGH WIND	WINTER STORM
3212.36	1420.3	1336.07	900.11	876.91	689.27	603.36	446.31	324.75	132.72

Analysis of Storm Data Set

Synopsis:

Data Processing:

Results: