In: Statistics and Probability
Let x be a random variable representing percentage change in neighborhood population in the past few years, and let y be a random variable representing crime rate (crimes per 1000 population). A random sample of six Denver neighborhoods gave the following information.
x | 27 | 3 | 11 | 17 | 7 | 6 |
y | 167 | 40 | 132 | 127 | 69 | 53 |
In this setting we have Σx = 71, Σy = 588, Σx2 = 1233, Σy2 = 70,612, and Σxy = 9041.
(e) For a neighborhood with x = 17% change in
population in the past few years, predict the change in the crime
rate (per 1000 residents). (Round your answer to one decimal
place.)
crimes per 1000 residents
(f) Find Se. (Round your answer to three
decimal places.)
Se =
(g) Find an 80% confidence interval for the change in crime rate
when the percentage change in population is x = 17%.
(Round your answers to one decimal place.)
lower limit | crimes per 1000 residents |
upper limit | crimes per 1000 residents |
(h) Test the claim that the slope β of the population
least-squares line is not zero at the 1% level of significance.
(Round your test statistic to three decimal places.)
t =
Find or estimate the P-value of the test statistic.
P-value > 0.2500.125 < P-value < 0.250 0.100 < P-value < 0.1250.075 < P-value < 0.1000.050 < P-value < 0.0750.025 < P-value < 0.0500.010 < P-value < 0.0250.005 < P-value < 0.0100.0005 < P-value < 0.005P-value < 0.0005
Conclusion
Reject the null hypothesis, there is sufficient evidence that β differs from 0.Reject the null hypothesis, there is insufficient evidence that β differs from 0. Fail to reject the null hypothesis, there is sufficient evidence that β differs from 0.Fail to reject the null hypothesis, there is insufficient evidence that β differs from 0.
(i) Find an 80% confidence interval for β and interpret
its meaning. (Round your answers to three decimal places.)
lower limit | |
upper limit |
Interpretation
For every percentage point increase in population, the crime rate per 1,000 increases by an amount that falls outside the confidence interval.For every percentage point decrease in population, the crime rate per 1,000 increases by an amount that falls within the confidence interval. For every percentage point increase in population, the crime rate per 1,000 increases by an amount that falls within the confidence interval.For every percentage point decrease in population, the crime rate per 1,000 increases by an amount that falls outside the confidence interval.
X | Y | XY | X² | Y² | |
total sum | 71.000 | 588.000 | 9041.000 | 1233 | 70612 |
sample size , n = 6
here, x̅ =Σx/n = 11.8333 , ȳ =
Σy/n = 98.00
SSxx = Σx² - (Σx)²/n = 392.833
SSxy= Σxy - (Σx*Σy)/n =
2083.000
SSyy = Σy²-(Σy)²/n = 12988.000
estimated slope , ß1 = SSxy/SSxx = 2083.000
/ 392.833 = 5.3025
intercept, ß0 = y̅-ß1* x̄ =
35.2537
so, regression line is Ŷ =
35.2537 + 5.3025 *x
----------------
e)Predicted Y at X= 17 is
Ŷ = 35.254 + 5.303
* 17 =
125.4
f)
SSE= (Sx*Sy - S²xy)/Sx = 1942.8859
std error ,Se = √(SSE/(n-2)) =
22.039
g)
Sample Size , n= 6
Degrees of Freedom,df=n-2 = 4
critical t Value=tα/2 = 1.533 [excel
function: =t.inv.2t(α/2,df) ]
X̅ = 11.83
Σ(x-x̅)² =Sxx 392.8
Standard Error of the Estimate,Se= 22.04
Predicted Y at X= 17 is
Ŷ = 35.254 + 5.303
* 17 = 125.396
standard error, S(ŷ)=Se*√(1/n+(X-X̅)²/Sxx) =
10.675
margin of error,E=t*Std error=t* S(ŷ) =
1.5332 * 10.6752 =
16.3673
Confidence Lower Limit=Ŷ +E =
125.396 - 16.3673 =
109.0
Confidence Upper Limit=Ŷ +E = 125.396
+ 16.3673 = 141.8
h)
estimated std error of slope =Se(ß1) = Se/√Sxx =
22.039 /√ 393 =
1.1120
t stat = estimated slope/std error =ß1 /Se(ß1) =
5.3025 / 1.1120
= 4.769
i)
confidence interval for slope
α= 0.2
t critical value= t α/2 =
1.533 [excel function: =t.inv.2t(α/2,df) ]
estimated std error of slope = Se/√Sxx =
22.03909 /√ 392.83
= 1.112
margin of error ,E= t*std error = 1.533
* 1.112 = 1.705
estimated slope , ß^ = 5.3025
lower confidence limit = estimated slope - margin of error
= 5.3025 - 1.705
= 3.598
upper confidence limit=estimated slope + margin of error
= 5.3025 + 1.705
= 7.007
For every percentage point decrease in population, the crime rate per 1,000 increases by an amount that falls within the confidence interval