---
title: "ANLY 512-Lab 2: WDI Indicator"
author: "Bo Li, Shan Huang, Langsheng Lin, Hao Zhang"
Date: "Aug 13 2017"
output: 
  flexdashboard::flex_dashboard:
    orientation: columns
    storyboard: true
    vertical_layout: fill
    social: menu
    source_code: embed

---

```{r setup, include=FALSE, message=FALSE}
library(dplyr)
library(reshape2)
library(ggplot2)
library(plotly)
library(ggthemes)
library(GGally)
library(gridExtra)

wdi <- read.csv('/Users/Katherine 1/Desktop/ANLY 512 Data Visualization/Lab2/WDI_csv/WDIData.csv', header = TRUE, stringsAsFactors = FALSE)
wdi <- as.data.frame(wdi)
colnames(wdi)[1] <- 'Country.Name'
colnames(wdi)[5:length(colnames(wdi))] <- gsub(pattern = 'X', replacement = '', x = colnames(wdi)[5:length(colnames(wdi))])
indicator <- c('GDP per capita growth (annual %)', 
               'Household final consumption expenditure per capita growth (annual %)',
               'Access to electricity (% of population)', 
               'Individuals using the Internet (% of population)',
               'Urban population growth (annual %)',
               'Renewable electricity output (% of total electricity output)',
               'Renewable energy consumption (% of total final energy consumption)')
country <- c('China', 'India', 'Russian Federation', 'South Africa', 'Brazil')

wdi.reduced <- wdi %>% 
                select(1,3,45:59) %>%
                filter(Indicator.Name %in% indicator, Country.Name %in% country) %>%
                melt(id = c('Country.Name', 'Indicator.Name'))
```

### Executive Summary

In this project, we analyzed the relationship among country's development level and people's life quality. 

The common macroeconomic theory indicated that people will benefit from country's continuous development, and it will in turn increast the development speed of the country. In this project, we choose five leading developing or newly industrialized countries, BRICS (Brazil, Russia, India, China, and South Africa), and performed empirical analysis to evaluate whether people's quality of life actually improved in thorough measurement metrics.

Through analysis, we conclude that:

1. All BRICS countries have enjoyed continuous growth, but 2008 global crisis made this growth much slower, and for some countries, the growth nearly stopped.

2. People's life quality improved a lot with the growth of each country. Hence it is consistent with our testing hypothesis that people will benefit from country's development for having better welfare and more convenient life.

3. Even after 2008 global financial crisis, the growth was adversely affected, however, the improvement of people's life in each countries didn't stop. 

4. For some BRICS countries, although they are still categoried as developing countries, people's life quality improved a lot during these 14 years in terms of several welfare indicators, such as the electricity access.

5. However, there is still room for further improvement compared with other developed countries. For instance, only around half of the population has access to Internet for four out of five BRICS countries. It may be due to the large population base of those countries, however, it is still an indicator that continuous improvement is needed to improve people's life quality.


### GDP per capita growth (annual %)

```{r, message=FALSE, warning=FALSE,fig.width = 20, fig.height = 12}
# The first plot type: plot single country and combine them together
# The function is used to draw line plot for a single country for a single indicator
# country_name:country name that wants to be plotted; indicator_name: indicator that wants to be plotted
single_line <- function (indicator_name, country_name) {
  temp <- wdi.reduced %>%
            filter(Indicator.Name == indicator_name, Country.Name == country_name)
  min_x <- min(as.numeric(as.character(temp$variable)))
  max_x <- max(as.numeric(as.character(temp$variable)))
  title_name <- paste(temp$Indicator.Name[1], '-', temp$Country.Name[1])
  
  p <- ggplot(temp, aes(x = as.numeric(levels(variable))[variable], y = value)) +
    geom_point() +
    geom_line() +
    labs(title = title_name, x = 'year', y = '%') +
    scale_x_continuous(breaks = round(seq(min_x, max_x, by = 1),0)) +
    theme_economist()  
  p
}

chn_gdp <- single_line('GDP per capita growth (annual %)', 'China')
ind_gdp <- single_line('GDP per capita growth (annual %)', 'India')
rus_gdp <- single_line('GDP per capita growth (annual %)', 'Russian Federation')
saf_gdp <- single_line('GDP per capita growth (annual %)', 'South Africa')
bra_gdp <- single_line('GDP per capita growth (annual %)', 'Brazil')

grid.arrange(chn_gdp,ind_gdp, rus_gdp, saf_gdp, bra_gdp, ncol = 2, nrow = 3)
```

***

Data Visualization Result and Conclusion- GDP per capita growth (annual %)

- All BRICS countries have enjoyed continuous growth in these 14 years. However, after global financial crisis from 2008, the growth rate was slower year by year.

- Since all countries have enjoyed a high-speed growth before crisis, it is expected that people's life quality and walfare will improve accordingly. After the crisis, because economic development was slower and the global economical enviornment was getting worse, it was expected that people's life quality improvement may be affected.


### Household final consumption expenditure per capita growth (annual %)

```{r, message=FALSE, warning=FALSE}
# The second plot type: using facet_grid feature in ggplot2 package
temp2 <- wdi.reduced %>%
          filter(Indicator.Name == 'Household final consumption expenditure per capita growth (annual %)')
min_x <- min(as.numeric(as.character(temp2$variable)))
max_x <- max(as.numeric(as.character(temp2$variable)))

q <- ggplot(temp2, aes(x = as.numeric(levels(variable))[variable], y = value)) + 
  geom_bar(stat="identity") +
  #geom_line() +
  labs(title = temp2$Indicator.Name[1], x = 'year', y = '%') +
  scale_x_continuous(breaks = round(seq(min_x, max_x, by = 1),0)) +
  facet_grid(Country.Name~.) +
  theme_economist()
ggplotly(q)
```

***

Data Visualization Result and Conclusion- Household final consumption expenditure per capita growth (annual %)

- Household final consumption is an effective indicator measuring the purchasing power of each family, and it is connected closely with macroenomic environment. It is expected that the development of the country will lead to higher purchasing power of its citizen. The comparison of pre-crisis level (2005-2008) with the beginning stage (around 2000) can indicate that expected trend, since the purchasing level has increased for all countries.

- Since all countries are affected by the crisis, it is expected that this will influence people's purchasing power as well. The figure confirmed it as well, since growth rate decreased a lot after 2008, and even negative increase was found for Russia and South Africa in 2009.

### Access to electricity (% of population)

```{r, message=FALSE, warning=FALSE}
temp3 <- wdi.reduced %>%
          filter(Indicator.Name == 'Access to electricity (% of population)')
min_x <- min(as.numeric(as.character(temp3$variable)))
max_x <- max(as.numeric(as.character(temp3$variable)))

q <- ggplot(temp3, aes(x = as.numeric(levels(variable))[variable], y = value)) + 
  geom_bar(stat="identity") +
  #geom_line() +
  labs(title = temp3$Indicator.Name[1], x = 'year', y = '%') +
  scale_x_continuous(breaks = round(seq(min_x, max_x, by = 1),0)) +
  facet_grid(Country.Name~.) +
  theme_economist()
ggplotly(q)
```

***

Data Visualization Result and Conclusion- Access to electricity (% of population)

- Over the 14-year time frame, the five countries have contiously increasing percentage of population with acess to electricity. 

- Percentage of Brazil, China and Russia population with access to electricity reaches 100% at this point. 

- However, there is a small portion of India and South Africa population without access to electricity. This percentage of India population is higher than that of South Africa. 

- Population with access to electricity provides evidence that people's living environment gets better and better. More and more people can enjoy the convenience. On the other hand, from industry standpoint, higher electricity access rate may indicate relatively high production, which is also reflected on GDP growth as show in previous figures.

### Individuals using the Internet (% of population)

```{r, message=FALSE, warning=FALSE}
temp4 <- wdi.reduced %>%
          filter(Indicator.Name == 'Individuals using the Internet (% of population)')
min_x <- min(as.numeric(as.character(temp4$variable)))
max_x <- max(as.numeric(as.character(temp4$variable)))

q <- ggplot(temp4, aes(x = as.numeric(levels(variable))[variable], y = value)) + 
  geom_bar(stat="identity") +
  #geom_line() +
  labs(title = temp4$Indicator.Name[1], x = 'year', y = '%') +
  scale_x_continuous(breaks = round(seq(min_x, max_x, by = 1),0)) +
  facet_grid(Country.Name~.) +
  theme_economist()
ggplotly(q)
```

***

Data Visualization Result and Conclusion- Individuals using the Internet (% of population)

- Percentage of individuals using the Internet is continously increasing. 

- However, only Russia goes above 60% while all the rest four countries fall below 60%.

- India has lowest rate and grows slowly. 

- This graph shows similiar trend among same country. Lowest rate of individuals using the Internet may be related to lowest rate of access to electricity compared to other BRICS countris. 

- 2010 is a key transition point when the Internet usage starts to grow rapidly for Russia and South Africa. In 2010, the GDP growth is much higher than that in previous years. Brazil, China and India remain steady growth during the 14 years.

- Overall, Internet usage rate has been increasing since 2000. The higher ratio indicates that more people live in a much moe convenient life with the access to the Internet, and it also indicates more technologies involved in each country's development strategy.

### Renewable electricity output (% of total electricity output) and Renewable energy consumption (% of total final energy consumption)

```{r, message=FALSE, warning=FALSE, fig.width = 24, fig.height = 12}
temp5 <- wdi.reduced %>%
          filter(Indicator.Name == 'Renewable electricity output (% of total electricity output)')
min_x <- min(as.numeric(as.character(temp5$variable)))
max_x <- max(as.numeric(as.character(temp5$variable)))

q1 <- ggplot(temp5, aes(x = as.numeric(levels(variable))[variable], y = value)) + 
        geom_bar(stat="identity") +
        #geom_line() +
        labs(title = temp5$Indicator.Name[1], x = 'year', y = '%') +
        scale_x_continuous(breaks = round(seq(min_x, max_x, by = 1),0)) +
        facet_grid(Country.Name~.) +
        theme_economist()

temp6 <- wdi.reduced %>%
          filter(Indicator.Name == 'Renewable energy consumption (% of total final energy consumption)')
min_x <- min(as.numeric(as.character(temp6$variable)))
max_x <- max(as.numeric(as.character(temp6$variable)))

q2 <- ggplot(temp6, aes(x = as.numeric(levels(variable))[variable], y = value)) + 
        geom_bar(stat="identity") +
        #geom_line() +
        labs(title = temp6$Indicator.Name[1], x = 'year', y = '%') +
        scale_x_continuous(breaks = round(seq(min_x, max_x, by = 1),0)) +
        facet_grid(Country.Name~.) +
        theme_economist()

grid.arrange(q1, q2, ncol = 2, nrow = 1)
```

***

Data Visualization Result and Conclusion- Renewable electricity output (% of total electricity output) and Renewable energy consumption (% of total final energy consumption)

-	Brazil has the highest renewable energy output percentage and consumption percentage due to its abundant natural resources.

-	China's both renewable electricity output percentage and consumption prcentage actually decreased over the tow decades, which explains the pollution situation in China.

-	India's renewable energy output percentage has slightly increased but its consumption percentage decreased significantly, which implies the similar problem with China that the development comes at the cost of environment.

-	Russia's both renewable energy output percentage and consumption percentage have been relatively the same over the two decades, indicating that the investment in renewable energy is proportional to the growth.

-	South Africa barely has any renewable energy output but has been able to utilize a steady percentage the renewable energy. 

### Urban population growth (annual %)

```{r, message=FALSE, warning=FALSE}
temp7 <- wdi.reduced %>%
          filter(Indicator.Name == 'Urban population growth (annual %)')
min_x <- min(as.numeric(as.character(temp7$variable)))
max_x <- max(as.numeric(as.character(temp7$variable)))

q7 <- ggplot(temp7, aes(x = as.numeric(levels(variable))[variable], y = value)) + 
        geom_line(stat="identity") +
        geom_point() +
        labs(title = temp7$Indicator.Name[1], x = 'year', y = '%') +
        scale_x_continuous(breaks = round(seq(min_x, max_x, by = 1),0)) +
        facet_grid(Country.Name~.) +
        theme_economist()

ggplotly(q7)
```

***

Data Visualization Result and Conclusion-Urban population growth (annual %)

- Urban population growth is an important indicator for a country's economic development. With a booming economy, the urban population growth rate is usually decreasing due to the increasing urban population.

- Since 2002, the urban population growth in Brazil, China, and India have continuiously decreased, While for Russia and South Afica, the rate has been increasing

- The difference in urban population growth rate might be due to the significant difference in each country's total population. Brazil, China, and India have a much larger population base than Russia and South Africa.

-Although the population growth is declining, the number of population can still increase significantly.