STA553-HW3

1. Reading the datasets and Reshaping

library(gapminder)
library(gganimate)
library(gifski)



income_per_person <- read.csv("https://raw.githubusercontent.com/rk2991/sta553/main/data/income_per_person.csv")
life_expectancy <- read.csv("https://raw.githubusercontent.com/rk2991/sta553/main/data/life_expectancy_years.csv")
country_region <- read.csv("https://raw.githubusercontent.com/rk2991/sta553/main/data/countries_total.csv")
population <- read.csv("https://raw.githubusercontent.com/rk2991/sta553/main/data/population_total.csv")

Reshaping the dataset from wide to long and then merginf the tables together to get the final output

# Reshape data set: Income Per Person to make a longitudinal data such that the resulting data set has three columns: country, year, and income.

income_long <- income_per_person %>% 
  mutate(Country=geo)%>% 
  gather(key = "Year", value = "Income", -geo , -Country , na.rm=TRUE) %>%
  mutate(year = as.numeric(substr(Year,2,5))) %>%
  select(-Year,-geo)
head(income_long)

# Do the same for Life Expectancy in Years so that the resulting data set has three columns: country, year, and life expectancy.

life_expectancy_long <- life_expectancy %>% 
   mutate(Country=geo)%>% 
  gather(key = "Year", value = "lifeExp", -geo, -Country, na.rm = TRUE) %>%
  mutate(year = as.numeric(substr(Year,2,5))) %>%
  select(-Year,-geo)
head(life_expectancy_long)

# combine the two tables together

LifeExpIncom <- merge(income_long, life_expectancy_long, by= c('Country','year'))
head(LifeExpIncom)

# Merge LifeExpIncom with country region so that the final data set has information about income, life expectancy, and country region

LifeExpIncomRegion <- merge(LifeExpIncom, country_region, by.x = "Country", by.y = "name", all.x = FALSE)
LifeExpIncomRegion <- select(LifeExpIncomRegion, year, Country, region, Income, lifeExp)
head(LifeExpIncomRegion)

# Reshaping Population data set
population_long <- population %>% 
  mutate(Country=geo)%>% 
  gather(key = "Year", value = "Population", -geo , -Country ,na.rm = TRUE) %>%
  mutate(year = as.numeric(substr(Year,2,5))) %>%
  select(-Year,-geo)


final_merge <- merge(LifeExpIncomRegion, population_long, by = c("Country", "year"), all.x = FALSE)
head(final_merge)

2. Interactive ScatterPlot

Interactive scatter plot to display the association between life expectancy and income for the year 2015.Following instructions are followed:

• Set the point size to be proportional to the population size
• Use different colors for different countries.
• Choose an appropriate transparency level so that overlapped points can be viewed.
• Choose an appropriate color to highlight the point boundary so that partially overlapped points can be easily distinguished.
• Include the country name and population size in the hover text.

country_data_2015 <- final_merge %>%
  filter(year == 2015)%>% 
  mutate(ppm = Population / 1000000) %>% 
select(Country, year, lifeExp, Income, Population, ppm)
head(country_data_2015)

plot_ly(
    data = country_data_2015,
    x = ~Income,  # Horizontal axis 
    y = ~lifeExp,   # Vertical axis 
    color = ~factor(Country),  # must be a numeric factor
    stroke = I("black"),
    hoverinfo = "text",
  hovertext = paste("Country: ", country_data_2015$Country,
                    "<br> Population: ", country_data_2015$ppm, "Million"),
  

     alpha  = 0.8,
     size = ~ppm,
   sizes = c(50, 1000),
     type = "scatter",
     mode = "markers" )%>% 
  layout(title = "Association Between Life Expectancy & Income in 2015",
                  titlefont = list(
                  size = 20,
                  color = 'darkred'),
         margin = 8,
          plot_bgcolor = "#E2E2E2", 
         xaxis = list(title = 'Income (US$)',erolinewidth = 1.5,
                              gridcolor = 'white'), 
         yaxis = list(title = 'Life Expectancy (Yr)',zerolinewidth = 1.5,
                              gridcolor = 'white'),
         showlegend = FALSE)

3. Animated ScatterPlot

Animated scatter plot showing pattern of change in the relationship between life expectancy and income over the years (considered the years from 1940 to 2018).The following instructions are followed:

• Set the point size to be proportional to the population size
• Use different colors for different regions.
• Choose an appropriate transparency level so that overlapped points can be viewed.
• Choose an appropriate color to highlight the point boundary so that partially overlapped points can be easily distinguished.

4. Interactive Map of Gas station

The following instructions are followed:

• Reading the gas station data set and taking a random sample of 500 gas stations from the US to plot on the map
• Included following information in the hover/popups: State, county, address and the zip code.

poc <- read_csv("https://raw.githubusercontent.com/rk2991/sta553/main/data/POC.csv")
poc_Sample <- poc[sample(nrow(poc), 500), ]


# geo styling
geostyle <- list(scope = 'usa',
                 projection = list(type = 'albers usa'),
                 showland = TRUE,
                 landcolor = toRGB("gray95"),
                 subunitcolor = toRGB("gray85"),
                 countrycolor = toRGB("gray85"),
                 countrywidth = 0.5,
                 subunitwidth = 0.5
               )
## plotting map
fig <- plot_geo(poc_Sample, lat = ~ycoord, lon = ~xcoord) %>%
       add_markers( text = ~paste("<br><b>State: </b>", STATE,
                                   "<br><b>County: </b>", county,
                                   "<br><b>Address: </b>", ADDRESS,
                                   "<br><b>Zip code: </b>", ZIPnew), 
                   color = "red", 
                   symbol = I("circle"), 
                   size = I(8), 
                   hoverinfo = "text" )   %>%
     layout( title = list( text = "500 Randomly Gas stations in the United States",
                              font = list( size = 18,
                                           color = "darkred" )),
                geo = geostyle )

fig