Task for Statistics assignment, linear regression analysis

davidius

New member
Joined
Feb 14, 2020
Messages
1
Dear people,

I am sorry to bother but I have a task to fulfill my assignment in statistic and I don't know how to solve it. Unfortunately the class got canceled several times but this assignment makes 40% of my grade therefore I need some help.

the task is as follows:

An MA student estimates the following regression model with data by district in Mexico City that she has obtained from Mexico City’s police records:

crime rate = α + β*deployment + ε,

where deployment measures the number of police officers deployed by district.

She finds a positive estimate for parameter β and concludes that more policing raises crime rates. However, her supervisor is not convinced about the estimate of parameter β reflecting the true effect of policing on crime rates.

1. What are the potential reasons for the supervisor’s skepticism?

2. In view of these reasons, describe a research strategy that, upon implementation, would deliver more convincing results on the cause-effect link between policing and crime rates.

(Answers to PART 2 are NOT supposed to be a general discussion. You are expected to produce answers that respond to the described situation and propose a concrete research strategy that improves upon the student’s results.)


For task 1: I simplified the regression as y = a + ß*x

I further had the idea to derive the "β" as in (y-a)/x = ß, but then I was unsure how to prove the point.


Sorry for inconveniences but I would be glad if someone could help me out.

Thanks a lot!
David
 
Top