Data 607 Project 2 downstream analysis
Christopher Estevez
Instructions
Data Loading
url1="https://raw.githubusercontent.com/chrisestevez/DataAnalytics/master/Project2/201606-citibike-tripdata.csv"
url2="https://raw.githubusercontent.com/chrisestevez/DataAnalytics/master/Project2/Crosstab%20Query.csv"
url3="https://raw.githubusercontent.com/chrisestevez/DataAnalytics/master/Project2/LendingClubData.csv"
urlBike=getURL(url1)
urlCross=getURL(url2)
urlLend=getURL(url3)
CitiBike = read.csv(text =urlBike,header = TRUE,stringsAsFactors = F,sep=",")
CrossTab = read.csv(text=urlCross,header = TRUE,stringsAsFactors = F,sep=",")
LendingTree = read.csv(text=urlLend,header = TRUE,stringsAsFactors = F,sep=",")Data Set 1
Compare monthly citizenship for the given regions.
Regions 2, 3 and 5 had the greatest population by month. This is also supported in the total population by region.
str(CrossTab)## 'data.frame': 9 obs. of 6 variables:
## $ Month : chr "April" "May" "June" "July" ...
## $ Region1: int 13 17 8 13 18 25 9 2 1
## $ Region2: int 33 55 63 104 121 160 88 86 128
## $ Region3: int 76 209 221 240 274 239 295 292 232
## $ Region4: int 2 1 1 6 9 2 2 2 6
## $ Region5: int 47 143 127 123 111 88 127 120 155Crossdata = gather(CrossTab,"Regions","Population",2:6)
ggplot(Crossdata, aes(Regions, Population,fill = Regions) ) +
geom_bar(stat="identity", width = 0.5)+
facet_grid(~ Month )+
theme(axis.text.x=element_text(angle=90,hjust=1,vjust=0.5))+
ggtitle("Population By Month")
Crossdata = Crossdata %>% group_by(Regions) %>% summarise(Total=sum(Population)) %>% arrange(desc(Total))
ggplot(Crossdata, aes(Regions, Total,fill = Regions) ) +
geom_bar(stat="identity", width = 0.5)+
theme(axis.text.x=element_text(angle=90,hjust=1,vjust=0.5))+
ggtitle("Total Population By Region")
Data set 2
NYC Citibike data
Intructions
The data has some missing values that will have to be addressed. The data can be analyzed to create a profile of they typical Citibike users/subscriber. The year can be converted to an age and converted to a categorical variable. Start and stop times can be used to calculate the average time each bike is used.
After subsetting and analyzing the data the findings were the following:
- The majority of the users for citibike from the sample data are male 1/3 are female.
- The majority of the usage falls from from 22 - 40 years old.
- The majority of the time spent on a city bike ranges 2-20 minutes.
str(CitiBike)## 'data.frame': 65499 obs. of 15 variables:
## $ tripduration : int 1470 229 344 1120 229 946 2351 773 1929 725 ...
## $ starttime : chr "6/1/2016 00:00:18" "6/1/2016 00:00:20" "6/1/2016 00:00:21" "6/1/2016 00:00:28" ...
## $ stoptime : chr "6/1/2016 00:24:48" "6/1/2016 00:04:09" "6/1/2016 00:06:06" "6/1/2016 00:19:09" ...
## $ start.station.id : int 380 3092 449 522 335 503 533 492 525 2002 ...
## $ start.station.name : chr "W 4 St & 7 Ave S" "Berry St & N 8 St" "W 52 St & 9 Ave" "E 51 St & Lexington Ave" ...
## $ start.station.latitude : num 40.7 40.7 40.8 40.8 40.7 ...
## $ start.station.longitude: num -74 -74 -74 -74 -74 ...
## $ end.station.id : int 3236 3103 469 401 285 495 386 483 306 3083 ...
## $ end.station.name : chr "W 42 St & Dyer Ave" "N 11 St & Wythe Ave" "Broadway & W 53 St" "Allen St & Rivington St" ...
## $ end.station.latitude : num 40.8 40.7 40.8 40.7 40.7 ...
## $ end.station.longitude : num -74 -74 -74 -74 -74 ...
## $ bikeid : int 19859 16233 22397 16231 15400 25193 19538 17101 17802 21421 ...
## $ usertype : chr "Subscriber" "Subscriber" "Subscriber" "Subscriber" ...
## $ birth.year : int 1972 1967 1989 1991 1989 1974 1986 1986 1968 1971 ...
## $ gender : int 1 1 1 1 1 1 1 1 1 2 ...Citi=CitiBike %>% drop_na() %>%
select(tripduration,start.station.id,end.station.id,usertype,birth.year,gender)%>%
mutate(TripTimeMinutes=round(tripduration/60, digits=0),Age= 2016 - birth.year)
Citi$gender = c('1'="Male",'2'="Female")[ as.character(Citi$gender)]
ggplot(Citi, aes(gender, fill = gender) ) +
geom_bar( width = 0.5)+
ggtitle("Usage By Gender")
#https://www.r-bloggers.com/how-to-make-a-histogram-with-ggplot2/
ggplot(data=Citi, aes(Age)) +
geom_histogram(breaks=seq(18,80, by =2), col="red",aes(fill=..count..)) +
scale_fill_gradient("Count", low = "green", high = "red") +
ggtitle("Age Histogram") 
ggplot(data=Citi, aes(TripTimeMinutes)) +
geom_histogram(breaks=seq(1,75, by =1), col="red",aes(fill=..count..)) +
scale_fill_gradient("Count", low = "green", high = "red")+
ggtitle("Trip time Histogram") 
Data set 3
Lending Club Data
https://www.lendingclub.com/info/download-data.action Just follow the link, we can download the data for the loans that were rejected and loans that is issued (Most current is 2016Q2). We can analyses the relationship among amount requested loan amount, debt to income ratio and employment length. Or we can compare income with interest rates and the rating of loans etc.
Due to the size of the data set being 50MB I removed some columns to reduce the size. My analysis focused on loans made by grade. A loan with grade A is a lower interest loan and consider less risky. The majority of the loans distributed were A,B, and C.
str(LendingTree)## 'data.frame': 97854 obs. of 14 variables:
## $ loan_amnt : int 18000 9800 28000 20000 4900 19625 30000 21000 9000 20000 ...
## $ funded_amnt : int 18000 9800 28000 20000 4900 19625 30000 21000 9000 20000 ...
## $ term : chr " 60 months" " 36 months" " 60 months" " 36 months" ...
## $ int_rate : chr "13.49%" "14.49%" "15.59%" "16.99%" ...
## $ installment : num 414 337 675 713 160 ...
## $ grade : chr "C" "C" "C" "D" ...
## $ sub_grade : chr "C2" "C4" "C5" "D1" ...
## $ emp_title : chr "Supervisor" "Restaurant manager " "Account Representative" "Transition Manager" ...
## $ emp_length : chr "10+ years" "10+ years" "10+ years" "2 years" ...
## $ home_ownership: chr "MORTGAGE" "MORTGAGE" "MORTGAGE" "MORTGAGE" ...
## $ annual_inc : num 70000 48000 86000 71000 120000 45000 323000 80000 52500 130000 ...
## $ purpose : chr "credit_card" "credit_card" "debt_consolidation" "credit_card" ...
## $ title : chr "Credit card refinancing" "Credit card refinancing" "Debt consolidation" "Credit card refinancing" ...
## $ addr_state : chr "VA" "OH" "AZ" "MI" ...LendData = LendingTree %>%
select(loan_amnt,grade,addr_state) %>%
group_by(addr_state,grade) %>%
summarise(TotalLoanAMT=sum(loan_amnt),TotalLoans= n())
ggplot(LendData, aes(grade, TotalLoanAMT,fill = grade) ) +
geom_bar(stat="identity", width = 0.5)+
ggtitle("Total Loan disbursement AMT By Grade")
ggplot(LendData, aes(grade, TotalLoans,fill = grade) ) +
geom_bar(stat="identity", width = 0.5)+
ggtitle("Total Loan disbursement By Grade")