Ottawa weather vs. Ottawa JavaScript Meetup attendance: an R analysis

ottawa-weather-javascriptmeetup.R

library(weatherData)
library(dplyr)
library(ggplot2)
library(gridExtra)

# Ottawa International Airport (YOW) weather data
getWeatherForYear = function(year) {
    getWeatherForDate('YOW', 
                      start_date= paste(sep='', year, '-01-01'),
                      end_date = paste(sep='', year, '-12-31'),
                      opt_detailed = FALSE,
                      opt_all_columns = TRUE)
}

# Execute functions - get data
df_weather = rbind(getWeatherForYear(2012), 
                    getWeatherForYear(2013),
                    getWeatherForYear(2014),
                    getWeatherForDate('YOW', start_date='2015-01-01',
                             end_date = '2015-02-11',
                             opt_detailed = FALSE,
                             opt_all_columns = TRUE))

# Write and read weather data 
write.csv(df_weather, 'ottawa_weatherdata.csv', row.names = TRUE)

weather = read.csv('ottawa_weatherdata.csv')
weather$Date = as.POSIXct(weather$Date)

weather %>% sample_n(10) %>% select(Date, Min_TemperatureC, Mean_TemperatureC, Max_TemperatureC)

# Read Meetup data 
events = read.csv('ottawajs_events.csv')
events$date =  as.POSIXct(events$date) 

events %>% sample_n(10) %>% select(event, rsvps, date)

# Group average attendance event by month
by_month = events %>% 
    mutate(month = factor(format(date, "%B"), levels=month.name)) %>%
    group_by(month) %>%
    summarise(events = n(), 
              count = sum(rsvps)) %>%
    mutate(avg = count / events) %>%
    arrange(desc(avg))

# Group average temperature by month
averagetemperature_bymonth = weather %>% 
    mutate(month = factor(format(Date, "%B"), levels=month.name)) %>%
    group_by(month) %>% 
    summarise(avg_temperature = mean(Mean_TemperatureC))

plot_temperature = ggplot(aes(x = month, y = avg_temperature, group=1), data = averagetemperature_bymonth) + 
    geom_line( ) + ggtitle("Average Temperature by Month")

plot_attendance = ggplot(aes(x = month, y = count, group=1), data = by_month) + 
    geom_bar(stat="identity", fill="grey50") +
    ggtitle("Average Attendance by Month")

grid.arrange(plot_temperature, plot_attendance, ncol = 1)

# Group by day 
by_day = events %>% 
    mutate(day = (as.POSIXct(events$date))) %>%
    group_by(day) %>%
    summarise(events = n(), 
              count = sum(rsvps)) %>%
    mutate(avg = count / events) %>%
    arrange(desc(avg))
weather = weather %>% mutate(day = Date)
merged = merge(weather, by_day, by = 'day')

ggplot(aes(x = count, y = Mean_TemperatureC, group = day), data = merged) + geom_point()

# Any correlation between the temperature and attendance?
cor(merged$count, merged$Mean_TemperatureC)