Reproducible Research: Assignment 2

1. Loading the database

First, the database has been loaded as below:

setwd("C:/Users/maxim/Desktop")
storm_data <- read.csv("repdata_data_StormData.csv.bz2")
stormdata<- storm_data[c("EVTYPE", "FATALITIES", "INJURIES", "PROPDMG", "PROPDMGEXP", "CROPDMG", "CROPDMGEXP")]

2. Processing the data

Please find attached two files to understand how the database has been written:

• National Weather Service Storm Data Documentation
• National Climatic Data Center Storm Events FAQ

We are interested in seven information:

• EVTYPE: the type of weather event
• FATALITIES: the number of deaths due to an event
• INJURIES: the number of injuries due to an event
• PROPDMG: the mantissa of the value of property damages (dollars)
• PROPDMGEXP: the exponent of the value of property damages (dollars). Only characters so it should be converted
• CROPDMG: the mantissa of the value of crops damages (dollars)
• CROPDMGEXP: the exponent of the value of property damages. Only characters so it should be converted

Then, the data has been processed as below:

stormdata<- storm_data[c("EVTYPE", "FATALITIES", "INJURIES", "PROPDMG", "PROPDMGEXP", "CROPDMG", "CROPDMGEXP")]

#Converting the exponent
transform_exposant <- function(e) {
  # h -> hundred, k -> thousand, m -> million, b -> billion
  if (e %in% c('h', 'H'))
    return(2)
  else if (e %in% c('k', 'K'))
    return(3)
  else if (e %in% c('m', 'M'))
    return(6)
  else if (e %in% c('b', 'B'))
    return(9)
  else if (!is.na(as.numeric(e))) # if a digit
    return(as.numeric(e))
  else if (e %in% c('', '-', '?', '+'))
    return(1)
  else {
    stop("Invalid exponent value.")
  }
}

property_damage <- sapply(stormdata$PROPDMGEXP, FUN=transform_exposant)
stormdata$PROPDMGEXP <- stormdata$PROPDMG * (10 ** property_damage)
crop_damage <- sapply(stormdata$CROPDMGEXP, FUN=transform_exposant)
stormdata$CROPDMGEXP <- stormdata$CROPDMG * (10 ** crop_damage)

Then, it is possible to get TOP15 of the most harmful events in the US from 1950 to 2011:

fatal_event <- aggregate(FATALITIES ~ EVTYPE, data=stormdata, sum)
injured_event <- aggregate(INJURIES ~ EVTYPE, data=stormdata, sum)

fatal_event_15 <- head(fatal_event[order(fatal_event$FATALITIES, decreasing = TRUE),],15)
injured_event_15 <- head(injured_event[order(injured_event$INJURIES, decreasing = TRUE),],15)

It is also possible to get TOP15 of the events which implied the highest costs in the US from 1950 to 2011:

property_damage_event <- aggregate(PROPDMGEXP ~ EVTYPE, data=stormdata, sum)
crop_damage_event <- aggregate(CROPDMGEXP ~ EVTYPE, data=stormdata, sum)

property_damage_event_15 <- head(property_damage_event[order(property_damage_event$PROPDMGEXP, decreasing = TRUE),],15)
crop_damage_event_15 <- head(crop_damage_event[order(crop_damage_event$CROPDMGEXP, decreasing = TRUE),],15)

1. Question 1

graph_fatalities <- ggplot(data=fatal_event_15, aes(x=reorder(EVTYPE, FATALITIES), y=FATALITIES, fill=FATALITIES)) +
  geom_bar(stat="identity") +
  coord_flip() +
  ylab("Total number of dead people") +
  xlab("Event type") +
  theme(legend.position="none")

graph_injuries <- ggplot(data=injured_event_15,aes(x=reorder(EVTYPE, INJURIES), y=INJURIES, fill=INJURIES)) +
  geom_bar(stat="identity") +
  coord_flip() + 
  ylab("Total number of injured people") +
  xlab("Event type") +
  theme(legend.position="none")


grid.arrange(graph_fatalities, graph_injuries, top=textGrob("Most harmful events in the US (1950-2011)",gp=gpar(fontsize=14,font=3)))

2. Question 2

graph_property_damage <- ggplot(data=property_damage_event_15,
             aes(x=reorder(EVTYPE, PROPDMGEXP), y=PROPDMGEXP, fill=PROPDMGEXP)) +
  geom_bar(stat="identity") +
  coord_flip() +
  ylab("Total number of property damages") +
  xlab("Event type") +
  theme(legend.position="none")

graph_crop_damage <- ggplot(data=crop_damage_event_15,
                                aes(x=reorder(EVTYPE, CROPDMGEXP), y=CROPDMGEXP, fill=CROPDMGEXP)) +
  geom_bar(stat="identity") +
  coord_flip() +
  ylab("Total number of crop damages") +
  xlab("Event type") +
  theme(legend.position="none")

grid.arrange(graph_property_damage, graph_crop_damage, top=textGrob("Events that caused greatest economic consequences in the US (1950-2011)",gp=gpar(fontsize=14,font=3)))