When you say "any robbery where a motor vehicle was stolen is considered a carjacking" are you defining car jacking as any time a person forcibly takes a car from the owner, or any time a car is stolen, including parked cars when the owner is not present?
One thing that would be interesting would be to see the mean and standard deviation of the delay between peak murder and peak carjacking, in months, using the rolling/moving average. If the data is too noisy when looking at individual agencies, perhaps the rough peak period, give or take two months, can be fitted to a Gaussian to extract a parametric peak time.
I was carjacked at gunpoint while driving a delivery van back in the 1970s and it was terrifying. Glad to see it is going down
When you say "any robbery where a motor vehicle was stolen is considered a carjacking" are you defining car jacking as any time a person forcibly takes a car from the owner, or any time a car is stolen, including parked cars when the owner is not present?
One mathematical comment: what you refer to as rolling average is called a moving average in signal processing (e.g. DSP).
The frequency transfer function is sinc (\omega T/2), times a T/2 delay (e^{-i \omega T/2}), where sinc(x)=sin(x)/x.
Edit: a bit rusty, both the delay and the sinc should have T/2, not T...
One thing that would be interesting would be to see the mean and standard deviation of the delay between peak murder and peak carjacking, in months, using the rolling/moving average. If the data is too noisy when looking at individual agencies, perhaps the rough peak period, give or take two months, can be fitted to a Gaussian to extract a parametric peak time.
Do these carjackings include those by ICE?
Fixed, thank you!