Data Manipulation with dplyr

# DATA: load data
data(storms)

# FILTER: keep rows that meet a condition
filter(storms, pressure < 1000)

# PIPE (%>%): send the result to the next step
# Shortcut: Ctrl + Shift + M
storms %>% filter(pressure < 1000)
topstorms <- storms %>% filter(pressure < 1000)
topstorms

# SELECT: keep or drop columns
topstorms %>% select(name)
topstorms %>% select(ends_with("diameter"))
topstorms %>% select(-ends_with("diameter"))
topstorms <- topstorms %>% select(-ends_with("diameter"))
topstorms

# GROUP_BY: group rows that share a value
topstorms %>% group_by(name)

# SUMMARIZE: one summary row per group
topstorms %>% 
  group_by(name) %>% 
  summarize(minpressure = min(pressure), 
            meanlat = mean(lat), 
            meanlong = mean(long))

topstorms %>% 
  group_by(name, year) %>% 
  summarize(count = n(), 
            minpressure = min(pressure), 
            meanlat = mean(lat), 
            meanlong = mean(long))

topstorms <- topstorms %>% 
  group_by(name, year) %>% 
  summarize(count = n(), 
            minpressure = min(pressure), 
            meanlat = mean(lat), 
            meanlong = mean(long))
topstorms

# MUTATE: add or change columns
topstorms %>% mutate(duration = 6 * count)
topstorms <- topstorms %>% mutate(duration = 6 * count)
topstorms

# ARRANGE: sort rows
topstorms %>% arrange(minpressure, desc(duration))
topstorms <- topstorms %>% arrange(minpressure, desc(duration))
topstorms

# WRITE.CSV: export data as a CSV file
write.csv(topstorms, "topstorms.csv")
write.csv(topstorms, "topstorms.csv", row.names = FALSE)

Technique: Converting data formats, Cleaning data
Tools: R
Data Format: Statistics