A national sporting good store wishes to use demographic
information to predict its monthly sales, in $1000s. Thrity-eight,
n=38, stores of the chain are randomly chosen across the
country. It is known that each store is approximately the same size
and carries the same merchandise.
The geographic area from which a store draws its customers is known
as the customer base. One of the variables is the percentage of the
customer base who have graduated from high school.
MonthlySalesi
= β0 +
β1PercentHSGradsi
+ ei
where,
MonthlySalesi
- is the total sales in month i, in $1000s
PercentHSGradsi
- is percentage of all customers in store i customer base
that have graduated from high school
A least-squares regression was ran in R producing the following
output:
Regression Analysis: MonthlySales versus PercentHSGrads
Predictor | Coef | SD Coef | T | P |
Constant | -2970 | 1371 | ||
PercentHSGrads | 59.66 | 17.67 |
S = 802.004 R-Sq =
Analysis of Variance | |||||
Source | DF | SS | MS | F | P |
Regression | 7333350 | ||||
Residual Error | 23155564 | ||||
Total | 37 | ||||
(g) A store located at a local mall has recently discovered that
90% of its customer base has a high school diploma. With 95%
confidence, estimate this store's monthly sales for the current
month.
Note: You will need
∑38i=1PercentHSGradsi=2935.17
and
∑38i=1(PercentHSGradsi)^2=228777
Lower Bound:
Upper Bound:
The following answers that are already on Chegg have turned out to be incorrect for the above question: (2398.7, 2400.1) , (1871.6, 2927.2)
Ʃx = 2935.17
Ʃx² = 228777
n = 38
x̅ = Ʃx/n = 2935.17/38 = 77.2413
SSxx = Ʃx² - (Ʃx)²/n = 228777 - (2935.17)²/38 = 2060.6071
Sum of Square error, SSE = 23155564
Standard error, se = √(SSE/(n-2)) = √(23155564/(38-2)) = 802.0038
Predicted value of y at x = 90
ŷ = -2970 + (59.66) * 90 = 2399.4
Significance level, α = 0.05
Critical value, t_c = T.INV.2T(0.05, 36) = 2.0281
g) 95% Prediction interval :
Lower limit = ŷ - tc*se*√(1 + (1/n) + ((x-x̅)²/(SSxx)))
= 2399.4 - 2.0281*802.0038*√(1 + (1/38) + ((90 - 77.2413)²/(2060.6071))) = 689.4
Upper limit = ŷ + tc*se*√(1 + (1/n) + ((x-x̅)²/(SSxx)))
= 2399.4 + 2.0281*802.0038*√(1 + (1/38) + ((90 - 77.2413)²/(2060.6071))) = 4109.4
Get Answers For Free
Most questions answered within 1 hours.