1. Local business owners suspect that crime in the area surrounding their stores
ID: 3255092 • Letter: 1
Question
1. Local business owners suspect that crime in the area surrounding their stores hurts their revenue. You are hired to conduct an expert analysis. You gather data on the number of misdemeanors (minor crimes "Crime and felonies (major crimes "Crimef that occur within one mile of each of 1,100 stores. You then regress each store's revenue Cin thousands of dollars) on the number of each type of crime that occurred within one mile of the store in the last year. You also differentiate revenue by the number of employees at the store ("Emp") and the square of the number of employees ("Emp Standard errors are presented in parentheses Rev 645.72 1.14Crime 3.78Crimes 10.74Emp 1.143 Emp (19.77) (0.44) (1.85) (8.25) (0.77) a) Interpret the coefficient on felony c in a sentence b) Test the hypothesis that felony crimes have no effect on revenue against the alternative that they do at the 95% confidence level c) One of the store owners claims that each felony that occurs in a neighborhood results in a loss of $6,000 dollars of revenue for stores. Test if you can reject his claim at the 90% confidence level d) What is the 95% confidence interval for the effect of misdemeanor crimes on store revenue? e) How large would the coefficient on employees ("Emp need to be in order for you to reject that it is equal to 0 at the 95% confidence level? You wish to test if felony crimes have the same effect on revenue as misdemeanor crimes (i.e. B1 B2). This is not possible with the current regression. Define a new variable and write a new regression equation that would make this comparison possible g) There were 5 felonies in the neighborhood of store A and 9 felonies in the neighborhood of store B. How many fewer misdemeanors would there need to be in the neighborhood of Store B in order for the two stores to have the same revenue h) You are concerned that whether or not the store is in an urban area is an omitted variable More felonies are committed in urban areas, but there are also more customers in urban areas so stores tend to earn more revenue. If you included the variable "Urban" in the regression, would the coefficient on felony crimes increase (get closer to 0) or decrease (become a large negative number)? ExplainExplanation / Answer
A) uNIT increase in felonies, the revenue goes down by 3.78 thousand dollars, keeping other variables unchanged.
b) Yes, Felonies is significant since p-value <0.05
calculation:
t-stat= coefficient/SE=2.04
p-value=T.DIST.2T(2.04,1100)=0.04
C) We need to compare how the current coefficient 3.78 is different from given 6
H0: coefficient is same (6 & 3.78); no significant difference between them
Ha: significant difference between coefficient
t-stat= (6-3.78)/1.85 =1.2
p-value=T.DIST.2T(1.2,1100)=0.23>.1
We cant reject H0. Thus, no significant difference between coefficients. His claim is acceptable.
d)
Excel formula:
t-stat= 1.962123 SE= 0.44 95% CI: Lower bound= 0.276666 Upper bound= 2.003334Related Questions
Navigate
Integrity-first tutoring: explanations and feedback only — we do not complete graded work. Learn more.