I need an explanation for this Economics question to help me study.
- The table below features three forecasting models used on the same set of data.
Model 1 |
Model 2 |
Model 3 |
|
Type |
Exponential Smoothing |
Regression |
Seasonal & Trend |
MSE |
8755.3 |
4876.2 |
5945.8 |
Based solely on the information in this output, which of the following is the best answer?(5)
- The data set contains no trend or seasonality.
- The data set contains trend but no seasonality.
- The data set contains seasonality but no trend.
- The data set probably contains cyclicality.
- The data set contains both trend and seasonality.
- In a forecasting application for 20 time periods, there are 10 negative errors and 10 positive errors.This indicates the model is performing well.(2)
- True
- False
- Refer to the following graph:
Which of the following apply?(8)
- The data contain a trend component.
- The data contain a seasonal component.
- The data ,contain a cyclical component.
- The data contain an irregular (random) component.
- In #3, which method (if any) is most appropriate?(4)
- Exponential smoothing.
- Regression.
- Regression with seasonal indices.
- None of the above.
- In #3, which of the following is most appropriate regarding sales?(4)
- We should use all of the data in our model.
- We should use only periods 5-16 in our model.
- We should use only periods 9-16 in our model.
- We should use only periods 13-16 in our model.
- We should use only periods 1-12 in our model.
- Refer to the Excel output on the final pages.Here, we are tracking the number of orders placed by week for a 20-week period.The first set of output is for an exponential smoothing model with α = 0.25.The second set of output is for a regression.Which of the following is most appropriate?(3)
- The exponential smoothing model is most appropriate.
- The regression is most appropriate.
- Another model would be more appropriate.
- The model with the lower MSE is always the most appropriate model.(2)
- True
- False
- In a given application, we are using regression with seasonal indices.The regression model is y = 42 + 2.5t.The seasonal indices for quarters 1-4 are 0.85, 0.92, 0.98, and 1.25, respectively.The predicted value for period 20 is ___________.(5)
- If our data contains seasonality but no trend, exponential smoothing is appropriate.(2)
- True
- False
- Annual data can exhibit seasonality.(2)
- True
- False
- We can assess quarterly seasonality with one year of data.(2)
- True
- False
Week |
Orders |
Forecast |
Error |
Error^2 |
1 |
45 |
#N/A |
#N/A |
|
2 |
56 |
45 |
11 |
121 |
3 |
65 |
47.75 |
17.25 |
297.5625 |
4 |
63 |
52.0625 |
10.9375 |
119.6289 |
5 |
54 |
54.79688 |
-0.79688 |
0.63501 |
6 |
60 |
54.59766 |
5.402344 |
29.18532 |
7 |
54 |
55.94824 |
-1.94824 |
3.795648 |
8 |
60 |
55.46118 |
4.538818 |
20.60087 |
9 |
56 |
56.59589 |
-0.59589 |
0.35508 |
10 |
57 |
56.44691 |
0.553085 |
0.305903 |
11 |
50 |
56.58519 |
-6.58519 |
43.36467 |
12 |
61 |
54.93889 |
6.06111 |
36.73706 |
13 |
47 |
56.45417 |
-9.45417 |
89.38128 |
14 |
56 |
54.09063 |
1.909375 |
3.645712 |
15 |
55 |
54.56797 |
0.432031 |
0.186651 |
16 |
52 |
54.67598 |
-2.67598 |
7.160852 |
17 |
57 |
54.00698 |
2.993017 |
8.958153 |
18 |
58 |
54.75524 |
3.244763 |
10.52849 |
19 |
61 |
55.56643 |
5.433572 |
29.52371 |
20 |
47 |
56.92482 |
-9.92482 |
98.50207 |
MSE = |
48.47673 |
SUMMARY OUTPUT |
|||||
Regression Statistics |
|||||
Multiple R |
0.139263 |
||||
R Square |
0.019394 |
||||
Adjusted R Square |
-0.03508 |
||||
Standard Error |
5.524367 |
||||
Observations |
20 |
||||
ANOVA |
|||||
|
df |
SS |
MS |
F |
Significance F |
Regression |
1 |
10.86466 |
10.86466 |
0.356001 |
0.558166112 |
Residual |
18 |
549.3353 |
30.51863 |
||
Total |
19 |
560.2 |
|||
|
Coefficients |
Standard Error |
t Stat |
P-value |
|
Intercept |
57.04211 |
2.566242 |
22.22787 |
1.54E-14 |
|
Week |
-0.12782 |
0.214226 |
-0.59666 |
0.558166 |