(18 points) Levitt (1997) estimates the relation between violent crime and numbe

Question

(18 points) Levitt (1997) estimates the relation between violent crime and number of police officers. He estimates: log(crime) = ?ˆ 0 ? 1.39 log(off icers) + 0.04 %femaleHH where log(crime) is the logged number of violent crimes per capita in a city, log(off icers) is the logged number of police officers per capita in a city, and %femaleHH is the percent of households in the city that are female-headed. (a) (6 points) Interpret the coefficient on log(off icers) (assuming we have ”good” estimates).

(b) (6 points) Interpret the coefficient on %femaleHH (assuming we have ”good” estimates).

(c) (6 points) Suppose that R2 = 0.212. Interpret this result (1-2 sentences).

Explanation / Answer

a) The 1.39 coefficient of Log(off icers ) interpretation is that for a unit increase in the logged number of police officers per capita in a city, there would be a 1.39 units increase in the logged number of violent crimes per capita in a city keeping all the other variables constant.

b) The 0.04 coefficient of the %female HH variable means that for a unit increase in the percent of households in the city that are female-headed, there would be a 0.04 increase in the logged number of violent crimes per capita in a city keeping all the other variables constant.

c) For a R2 = 0.212, the interpretation is that 21.2% of the total variation in the dependent variable that is the logged number of violent crimes per capita in a city keeping all the other variables constant, is explained by the regression equation here and rest is still the unexplained variation in the dependent variable.

Related Questions

Navigate

• Browse (All)
• Browse #
• Subjects

Integrity-first tutoring: explanations and feedback only — we do not complete graded work. Learn more.