This capstone project aims to explore one potential Uber/Lyft’s impact: whether daily Uber/Lyft trips affect parking violations. NYC daily Uber/Lyft trip and parking ticket data are collected and correlated by taxi cab zone. Three technical models, Fixed Effects, Difference in Difference, and Bayesian Network, are applied on the prepared data. The results of these models show the negative correlation and causal effect between the number of Uber/Lyft trips and parking tickets, suggesting Uber/Lyft help in reducing parking violations in NYC. Given the controversial issues around TNC, this capstone project can assisting in understanding impact of Uber/Lyft and offer policy insight to the TNC regulation. (Full report, PPT)
Usage
Step 1: Data Collection
Data Collection: Developed a pipeline to collect all the 42-month FHV and 6-year parking ticket datasets
Note: all data is open on the Internet. For every data, we've provided either the link or the intermediated form transformed by us. Also, feel free to change data pathes when re-using these codes, as the data may be called differently in our notebooks
Step 2: Data Preprocessing
Basic Geo-unit of all data: taxi zone
OBJECTID / LocationID
Shape_Leng
Shape_Area
borough
Taxi zone ID
Taxi zone length
Taxi zone area
NYC borough number
Uber & Lyft Data: Filtered and grouped the Uber/Lyft trips by taxi zones from FHV trips with Spark.
Tickets Data
Street Name: Filtered and grouped ticket data with Spark
Geocoding: Converted 350 thousand street names into coordinates through Google Geocoding API
Taxi Zone: Mapped the coordinates into taxi zones with R-tree method with Spark.
Additional Data
Temporal Data
Spatial Data
Potential Data
Weather, Holiday and Weekdays
ACS, Crime, Transportation, Education and Parking Facilities
Events, BBL, Park, Parking Regulation Locations and Signs, Meter parking price, Garage parking price, Google POI and Yelp
ACS details
DensityPop
IncomePerCap
Poverty
Professional
Service
Office
Population Density
Income per capita ($)
% under poverty level rate
% employed in management, business, science, and arts
% employed in service jobs
% employed in sales and office jobs
Production
Employed
Unemployment
Drive
Carpool
Transit
Walk
% employed in production, transportation, and material movement
% employed rate (16+)
% Unemployment rate
% commuting alone in a car, van, or truck
% carpooling in a car, van, or truck
% commuting on public transportation
% walking to work
OtherTransp
WorkAtHome
MeanCommuteMean
Construction
% commuting via other means
% working at home
commute time (minutes)
% employed in natural resources, construction, and maintenance
Crime details
FELONY
VIOLATION
MISDEMEANOR
Number of felony crimes in the taxi zone
Number of violation crimes in the taxi zone
Number of misdemeanor crimes in the taxi zone
Transportation, Parking Facilities and Education details