1) Profitable Industries

Let’s compare earnings and information figures between US economic industries to find which sector might be most worthy of our focus.

Insight: Power, Computers, and Telecommunications Services display disproportionately high EBITDA’s for their market sizes (number of firms). Therefore, those three sectors (two of which are connected to tech enabled services) may be of interest, as they can be considered somewhat undertapped yet profitable markets.

#Interactive scatterplot of EBITDA (in thousands of $) and number of firms in each industry


p <- data %>%
  ggplot(aes(EBITDA, nfirms, size = Revenues, color=Industry_Name)) +
  geom_point() +
  theme_bw() 
ggplotly(p)

2) Balance Sheets

On a more microeconomic scale, once we’ve chosen a company of interest, it is important to analyze its balance sheet.

Insight: Assets strongly trump liabilities here, meaning the company is worthy of more research and attention.

#Balance Sheet Pie Chart
favstats(amt~Category, data=data2)
##      Category   min      Q1 median       Q3    max     mean       sd  n missing
## 1     Assets     33  475.00   1297 18050.00 118104 18013.87 33348.36 31       1
## 2      Equity -3043 -186.00   5925 15817.25  29870  9699.00 13336.94  8       0
## 3 Liabilities     8  884.75   2390 18598.00 128577 21856.27 38672.51 22       0
lsum <- 21856.27    * 22
asum <- 18013.87 * 31
esum <- 9699.00 * 8
library(plotrix)
## 
## Attaching package: 'plotrix'
## The following object is masked from 'package:mosaic':
## 
##     rescale
sums <- c(lsum, asum, esum)
labels <- c("Liabilities", "Assets", "          Equities")
pie3D(sums, explode=0.1, labels = labels, main="Balance Sheet Distribution")

3) Mapping Sales Frequency

Because Vista Highland has surperior interns and leadership, it takes over a large share of Uber for a lower price, with Mike Drinkwater as the new CEO. A map of ride pickups allows for more efficient marketing, beneficial price discrimation, and more.

Insight: The map predictably displays high ride densities in NYC. However, many rides are requested in New Haven and Stamford, CT.

library(leaflet)
uber <- uber %>%
  filter(Lat < 42)
ubermap <- leaflet(uber) %>%
  addTiles() %>%
  addMarkers(lng = ~Lon,
                   lat = ~Lat, label = ~Date_Time) 
ubermap