Analysis

getwd()
tickets <- read.csv("tickets.csv", header = T)

Analysis of Daily Revenues

#Daily Analysis
DailyTotals<- tickets %>% group_by(Citation.Issue.Date) %>% summarise(Total = sum(Amount.Paid))

head(DailyTotals)

## # A tibble: 6 × 2
##   Citation.Issue.Date Total
##                <fctr> <int>
## 1          2009-01-11   313
## 2          2009-01-19    97
## 3          2009-02-01    39
## 4          2009-02-04   114
## 5          2009-02-10    89
## 6          2009-02-15     3

AverageDay <- mean(DailyTotals$Total)
AverageDay

## [1] 884.5944

MedianDay <- median(DailyTotals$Total)
MedianDay

## [1] 619

Daily <- DailyTotals$Total

var(Daily)

## [1] 799183.9

sd(Daily)

## [1] 893.9708

quantile(Daily, c(.1, .3, .5, .7, .9))

##    10%    30%    50%    70%    90% 
##   68.9  278.0  619.0 1100.0 2120.8

shapiro.test(Daily)

## 
##  Shapiro-Wilk normality test
## 
## data:  Daily
## W = 0.83609, p-value < 2.2e-16

#EDA Graphs for Rushhour
qqnorm(Daily)
qqline(Daily)

ggplot(DailyTotals, aes(x = Total)) + 
  geom_histogram(aes(y=..density..), colour="black", fill="pink") + 
  geom_density(alpha=.5, fill="red") + 
  geom_vline(xintercept = AverageDay, linetype = "longdash", colour = "blue") + geom_vline(xintercept = MedianDay, linetype = "twodash" , colour = "green")

## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.

Analysis of Weekly Revenues

#Create the week variable
tickets$Citation.Issue.Date <- as.Date(tickets$Citation.Issue.Date)
dateRange <- c(tickets$Citation.Issue.Date) 
x <- as.POSIXlt(dateRange) 
tickets$Citation.Issue.Week <- strftime(x,format="%W")

#Weekly Total Revenue Mean compared to each Week's Total Revenue
#Weekly Analysis
WeeklyTotals<- tickets %>% group_by(Citation.Issue.Week, Citation.Issue.Month, Citation.Issue.Year) %>% summarise(Total = sum(Amount.Paid))

head(WeeklyTotals)

## Source: local data frame [6 x 4]
## Groups: Citation.Issue.Week, Citation.Issue.Month [2]
## 
##   Citation.Issue.Week Citation.Issue.Month Citation.Issue.Year Total
##                 <chr>               <fctr>               <int> <int>
## 1                  00                  Jan                2010  1482
## 2                  00                  Jan                2011   247
## 3                  00                  Jan                2012    92
## 4                  00                  Jan                2013  3466
## 5                  00                  Jan                2014   789
## 6                  01                  Jan                2009   313

MeanWeek <- mean(WeeklyTotals$Total)
MeanWeek

## [1] 4059.011

MedianWeek <- median(WeeklyTotals$Total)
MedianWeek

## [1] 3512

Weekly <- WeeklyTotals$Total

var(Weekly)

## [1] 9399593

sd(Weekly)

## [1] 3065.876

quantile(Weekly, c(.1, .3, .5, .7, .9))

##    10%    30%    50%    70%    90% 
##  395.4 1888.4 3512.0 5428.0 8323.4

shapiro.test(Weekly)

## 
##  Shapiro-Wilk normality test
## 
## data:  Weekly
## W = 0.93628, p-value = 1.966e-11

qqnorm(Weekly)
qqline(Weekly)

ggplot(WeeklyTotals, aes(x = Total)) + geom_histogram(aes(y = ..density..), fill = "pink", colour = "black", binwidth = 500) + geom_density(alpha=.5, fill="red") + geom_vline(xintercept = MeanWeek, linetype = "longdash", colour = "blue") + geom_vline(xintercept = MedianWeek, linetype = "twodash" , colour = "green")

sd.Week <- sd(WeeklyTotals$Total)
sd.Week

## [1] 3065.876

Analysis of Monthly Revenues

#Monthly Analysis
MonthlyTotals<- tickets %>% group_by(Citation.Issue.Month, Citation.Issue.Year) %>% summarise(Total = sum(Amount.Paid))

head(MonthlyTotals)

## Source: local data frame [6 x 3]
## Groups: Citation.Issue.Month [1]
## 
##   Citation.Issue.Month Citation.Issue.Year Total
##                 <fctr>               <int> <int>
## 1                  Apr                2009 10409
## 2                  Apr                2010   812
## 3                  Apr                2011 23206
## 4                  Apr                2012 26792
## 5                  Apr                2013 31983
## 6                  Apr                2014 31477

#EDA of Statistics of Monthly Revenue
AverageMonth <- mean(MonthlyTotals$Total)
AverageMonth

## [1] 20689.68

MedianMonth <- median(MonthlyTotals$Total)
MedianMonth

## [1] 20284.5

Monthly <- MonthlyTotals$Total

var(Monthly)

## [1] 117427178

sd(Monthly)

## [1] 10836.38

quantile(Monthly, c(.1, .3, .5, .7, .9))

##     10%     30%     50%     70%     90% 
##  9058.8 13586.1 20284.5 26730.8 32311.2

shapiro.test(Monthly)

## 
##  Shapiro-Wilk normality test
## 
## data:  Monthly
## W = 0.97677, p-value = 0.2037

#EDA Graphs for Rushhour
qqnorm(Monthly)
qqline(Monthly)

ggplot(MonthlyTotals, aes(x = Total)) + geom_histogram(aes(y = ..density..), fill = "pink", colour = "black", binwidth = 2000) + geom_density(alpha=.5, fill= "red") + geom_vline(xintercept = AverageMonth, linetype = "longdash", colour = "blue") + geom_vline(xintercept = MedianMonth, linetype = "twodash" , colour = "green")