In: Statistics and Probability
Carolina Wood Products, Inc., a major manufacturer of household furniture, is interested in predicting expenditures on furniture (FURN) for the entire United States.
Period | FURN ($Billions) | Time |
Mar-07 | 61.1 | 1 |
Jun-07 | 59.8 | 2 |
Sep-07 | 59 | 3 |
Dec-07 | 58 | 4 |
Mar-08 | 56.2 | 5 |
Jun-08 | 58.1 | 6 |
Sep-08 | 59.2 | 7 |
Dec-08 | 61.4 | 8 |
Mar-09 | 63.7 | 9 |
Jun-09 | 67.4 | 10 |
Sep-09 | 71.1 | 11 |
Dec-09 | 74.1 | 12 |
Mar-10 | 77.3 | 13 |
Jun-10 | 80.2 | 14 |
Sep-10 | 82.4 | 15 |
Dec-10 | 85.7 | 16 |
Mar-11 | 88.9 | 17 |
Jun-11 | 92.3 | 18 |
Sep-11 | 95.2 | 19 |
Dec-11 | 99.6 | 20 |
Mar-12 | 100.4 | 21 |
Jun-12 | 104.4 | 22 |
Sep-12 | 108.3 | 23 |
Dec-12 | 110.7 | 24 |
Mar-13 | 111.8 | 25 |
Jun-13 | 113.2 | 26 |
Sep-13 | 116.4 | 27 |
Dec-13 | 117.2 | 28 |
Mar-14 | 122.8 | 29 |
Jun-14 | 127.4 | 30 |
Sep-14 | 129.2 | 31 |
Dec-14 | 132.7 | 32 |
Mar-15 | 136.7 | 33 |
Jun-15 | 138.5 | 34 |
Sep-15 | 138 | 35 |
Dec-15 | 138.7 | 36 |
Mar-16 | 144.4 | 37 |
Jun-16 | 143 | 38 |
Sep-16 | 142.7 | 39 |
Dec-16 | 139.3 | 40 |
Mar-17 | 41 | |
Jun-17 | 42 | |
Sep-17 | 43 | |
Dec-17 | 44 |
FURN = a + b(TIME)
Period | Time | Trend Forecast |
2017 Q1 | 41 | |
2017 Q2 | 42 | |
2017 Q3 | 43 | |
2017 Q4 | 44 |
a)
ΣX | ΣY | Σ(x-x̅)² | Σ(y-ȳ)² | Σ(x-x̅)(y-ȳ) | |
total sum | 820 | 3966.5 | 5330 | 36121.3 | 13746.35 |
mean | 20.50 | 99.16 | SSxx | SSyy | SSxy |
sample size , n = 40
here, x̅ = Σx / n= 20.50 ,
ȳ = Σy/n = 99.16
SSxx = Σ(x-x̅)² = 5330.0000
SSxy= Σ(x-x̅)(y-ȳ) = 13746.4
estimated slope , ß1 = SSxy/SSxx = 13746.4
/ 5330.000 = 2.5791
intercept, ß0 = y̅-ß1* x̄ =
46.2919
so, regression line is Ŷ =
46.2919 + 2.5791
*x
b)
SSE= (SSxx * SSyy - SS²xy)/SSxx =
668.715
std error ,Se = √(SSE/(n-2)) =
4.195
correlation coefficient , r = Sxy/√(Sx.Sy)
= 0.9907
R² = (Sxy)²/(Sx.Sy) = 0.9815
As we can see R square = 98.15 % means 98.15% variance in FURN can be explained by time period. Model is good.
C)
Predicted Y at X= 41 is
Ŷ = 46.29192 +
2.579053 * 41 =
152.033
Predicted Y at X= 42 is
Ŷ = 46.29192 +
2.579053 * 42 =
154.612
Predicted Y at X= 43 is
Ŷ = 46.29192 +
2.579053 * 43 =
157.191
Predicted Y at X= 44 is
Ŷ = 46.29192 +
2.579053 * 44 =
159.770
THANKS
revert back for doubt
please upvote