Before the adoption of Vision Zero, in 2012 Hurricane Sandy struck NYC and damaged the infrastructure of the street. Hurricane Sandy caused extensive impacts to infrastructure causing closures of all major transportation arteries into the city on October 31 to November 1, 2012. A potential reason that we see a slow progress of improvement with the adoption of Vision Zero, Hurricane Sandy might have been a contributing factor to the slow progress.  

Vision Zero Data: 

The data used for this analysis driven from:
NYC Open Data NYPD Motor Vehicle Collison. (https://data.cityofnewyork.us/api/views/h9gi-nx95/rows.csv?accessType=DOWNLOAD
  1. Filtered the data set to before and after the adoption of Vision Zero
  2. Used the cumulative number of accident per day 
  3. Using a time frame using Year, Month and Day (To create the time series plot) 
Used the following the link for understandings: 
https://www1.nyc.gov/site/visionzero/index.page
http://www.nyc.gov/html/dot/html/about/datafeeds.shtml#vision

Methodology:

To investigate the effectiveness of Vision Zero, a Kolmogorov-Smirnov (K-S) test is employed. KS-test tests whether two samples are drawn from the same distribution or if the two samples differ significantly from each other. The null hypothesis in the K-S statistic is that the distribution of the two samples is the same.
K-S test is employed for the following reasons:
  1. The test is distribution-free, thus giving valid probabilities for any underlying distribution of the two samples.
  2. It can be universally applied without restriction to any problem.
  3. Critical values of probabilities are widely available.
  4. The statistic is easy to compute.
Applying to the Vision Zero case, the Null hypothesis is that the distribution of the number of daily accidents before Vision Zero adoption and after Vision Zero adoption is the same. We reject the Null hypothesis if the resulting p-value from the test is less than the critical value 0.01.
point of change analysis is also conducted to assess if the Vision Zero implementation date corresponds to a point in time when there is a significant difference between the mean number of daily accidents before and after the point. This will help assess if the Vision Zero implementation date coincides with the point in time where the rate of daily accidents increases/decreases.