Citibike Report


This report aims to find whether mean trip duration of young people is longer than middle-aged people. Z-test is performed to determine whether this hypothesis is true. After performed the test, it's very likely that young people ride longer than middle-aged people.


The data source is citibikedata, which contains the information of people using citibikes. Columns of birth years and trip durations of the data are used. To classify young and middle-aged
people, people whose age is between 21 to 40 are defined as young and whose age is between 40 and 59 are defined as elder.


Null hypothesis test was performed. The null hypothesis is: Total trip duration of people who are in the age of (20,40] is equal or shorter than those in the age of (40,60] in the year 2015 in NYC. (with 5% significance)
The normalized total trip duration of 5 different age range is shown below. It shows each certain age range's total trip duration divided by all age ranges' total trip duration.

Below is the error bar with 5% significance level of the above chart. 

Since there are sufficient data, according to law of large number, I performed the Z-test.
By performing the Z-test, the Z value is 6.33>1.645, so that we can reject the null hypothesis.


The null hypothesis is false. According to the result of Z-test, mean trip duration of people who are in the age of (20,40] is longer than those in the age of (40,60]. (with 5% significance) in NYC in 2015.

[Someone else is editing this]

You are editing this file