Import and clean data.

require(dplyr)
## Loading required package: dplyr
## 
## Attaching package: 'dplyr'
## 
## The following objects are masked from 'package:stats':
## 
##     filter, lag
## 
## The following objects are masked from 'package:base':
## 
##     intersect, setdiff, setequal, union
require(tidyr)
## Loading required package: tidyr
digaddat <- read.csv("C:\\Users\\Andrew\\Desktop\\Cuny\\Data Acquisition\\Project 2\\Ex1Advertising\\digadscomp.csv")

colnames(digaddat)[1] <- "year"

Tidying data. A little easier than I expected.

digitalads <- digaddat %>%
  gather("company", "revenue", Google:AOL)

tbl_df(digitalads)
## Source: local data frame [25 x 3]
## 
##     year  company revenue
##    (int)   (fctr)   (dbl)
## 1   2009   Google    0.36
## 2   2010   Google    0.86
## 3   2011   Google    1.67
## 4   2012   Google    2.26
## 5   2013   Google    2.99
## 6   2009 Facebook    0.56
## 7   2010 Facebook    1.21
## 8   2011 Facebook    1.73
## 9   2012 Facebook    2.18
## 10  2013 Facebook    3.17
## ..   ...      ...     ...

Total, average, and change in revenue by company. Facebook and Google earn the most revenue and have the largest growth as well. Yahoo is still a heavy hitter, but with stunted growth.

digitalads %>%
  spread(year, revenue) %>%
  mutate(improverevenue = `2013` - `2009`) %>%
  gather(year, revenue, `2009`:`2013`) %>%
  group_by(company) %>%
  mutate(averagerevenue = mean(revenue)) %>%
  group_by(company, improverevenue, averagerevenue) %>%
  summarise(totalrevenue = sum(revenue)) %>%
  arrange(desc(totalrevenue))
## Source: local data frame [5 x 4]
## Groups: company, improverevenue [5]
## 
##     company improverevenue averagerevenue totalrevenue
##      (fctr)          (dbl)          (dbl)        (dbl)
## 1    Google           2.63          1.628         8.14
## 2  Facebook           2.61          1.770         8.85
## 3     Yahoo           0.01          1.334         6.67
## 4 Microsoft           0.42          0.634         3.17
## 5       AOL           0.22          0.588         2.94

Total and average revenue by year; steady yearly total and average growth overall.

digitalads %>%
  group_by(year) %>%
  mutate(totalrevenue = sum(revenue)) %>%
  group_by(year, totalrevenue) %>%
  summarise(averagerevenue = mean(revenue))
## Source: local data frame [5 x 3]
## Groups: year [?]
## 
##    year totalrevenue averagerevenue
##   (int)        (dbl)          (dbl)
## 1  2009         3.06          0.612
## 2  2010         4.48          0.896
## 3  2011         5.89          1.178
## 4  2012         7.39          1.478
## 5  2013         8.95          1.790