library(ggplot2)
library(plyr)
library(knitr)

data_path <- '/Users/ilanreinstein/Documents/beer/'
styles <- read.csv(paste(data_path, 'styles.csv', sep = ''),)
cats <- read.csv(paste(data_path, 'categories.csv', sep = ''),)
brew <- read.csv(paste(data_path, 'breweries.csv', sep = ''),)
beer <- read.csv(paste(data_path, 'beers.csv', sep = ''),)
cons <- read.csv(paste(data_path, 'consum.csv', sep = ''),)

Which countries consume more beer per capita?

cons$country <- factor(cons$country, levels = cons$country, ordered = T)
cons <- cons[order(cons$cons_lts),]
ggplot(cons, aes(country, cons_lts, fill = country)) + geom_bar(stat = 'identity', fill = 'skyblue', alpha =  0.8) + theme_bw() + theme(axis.text.x = element_text(angle = 90)) + labs(y = 'Consumption per capita (lts)')

Consumption per capita

What is the most common category?

A <- merge(beer, cats, by.x = 'cat_id', by.y = 'id')
A <- transform(A, cat_name = ordered(cat_name, levels = names(sort(-table(cat_name)))))
ggplot(A, aes(x = cat_name, fill = cat_name)) + geom_histogram() + theme_bw() + theme(axis.text.x = element_text(angle = 90), legend.title = element_blank()) + labs(x = 'Category', y = 'No. of brands') + scale_fill_brewer(palette = 'Spectral')

Common category

What is the mean abv (alcohol volume) by category?

A$abv[A$abv==0] <- NA
A$abv[A$abv > 50] <- NA
ABV <- ddply(A, .(cat_name), summarise, MeanABV = mean(abv,na.rm = T))
ABV1 <- ABV[order(ABV$MeanABV, decreasing = T),] 
colnames(ABV1)[1] <- 'Category'
kable(ABV1,row.names = F)

Category	MeanABV
Belgian and French Ale	7.389
British Ale	6.862
North American Ale	6.788
German Lager	6.220
Irish Ale	5.936
Other Style	5.827
German Ale	5.662
North American Lager	4.968
Other Lager	4.550

ggplot(A, aes(x = cat_name, y = abv, fill = cat_name)) + geom_boxplot() + theme_bw() + theme(axis.text.x = element_text(angle = 90), legend.title = element_blank()) + scale_fill_brewer(palette = 'Spectral') + labs(x = 'Category', y = 'Alcohol Volume')

## Warning: Removed 2858 rows containing non-finite values (stat_boxplot).

Alcohol Volume by category

How many styles are held by category?

B <- merge(cats, styles, by.x = 'id', by.y = 'cat_id')

## Warning: column name 'id' is duplicated in the result

B <- transform(B, cat_name = ordered(cat_name, levels = names(sort(-table(cat_name)))))
q <- ggplot(B, aes(x = cat_name, fill = cat_name)) + geom_histogram() + theme_bw() + theme(legend.title = element_blank()) + labs(y = 'No. of styles', x = 'Category') + scale_fill_brewer(palette = 'Spectral')
q <- q + coord_polar()
q <- q + theme(
              axis.line=element_blank()
              ,axis.ticks=element_blank()
              ,axis.title.x=element_blank()
              ,axis.title.y=element_blank()
              ,panel.border=element_blank()
  )  
q

Styles by category