For this problem use the data in the worksheet named below.

The data shows the "quarterly" energy bills for a school system in units of $1000.

Answer the following questions.

t	Quarter	y
1	1	344.39
2	2	246.63
3	3	131.53
4	4	298.87
5	1	313.45
6	2	189.76
7	3	179.10
8	4	221.10
9	1	246.84
10	2	209.00
11	3	51.21
12	4	133.89
13	1	277.01
14	2	197.98
15	3	50.68
16	4	218.08
17	1	365.10
18	2	207.51
19	3	54.63
20	4	214.09
21	1	267.00
22	2	230.28
23	3	230.32
24	4	436.41
25	1	467.06
26	2	306.03
27	3	253.23
28	4	279.46
29	1	336.56
30	2	196.67
31	3	152.15
32	4	319.67
33	1	440.00
34	2	315.04
35	3	216.42
36	4	339.78
37	1	434.66
38	2	399.66
39	3	330.80
40	4	539.78

1. Should we keep all the dummy variables in the model, even though only one or two are significant?

Group of answer choices

We use the insignificant quarter as the base and introduce three new dummy variables

Yes, but only keep the ones that are significant

Yes, we keep them all

No, we exclude them all

To continue with the questions above, assume we get the following values for the coefficients. You can copy and paste these values to your Excel file and use them instead of the regression coefficients you have obtained.

Coefficient	Value
Intercept	279
t	-8
t^2	0.3
Q1	64
Q2	-40
Q3	-130

1.Forecast the energy expenditure for the first quarter of the following year (period 41).

Group of answer choices

519.30 (x$1000)

563.30 (x$1000)

522.30 (x$1000)

349.30 (x$1000)

2. Forecast the energy expenditure for the second quarter of the following year (period 42).

Group of answer choices

303.00 (x $1000)

436.20 (x $1000)

473.20 (x $1000)

432.20 (x $1000)

3.Forecast the energy expenditure for the fourth quarter of the following year (period 44).

Group of answer choices

550.80 (x$1000)

507.80 (x$1000)

511.80 (x$1000)

471.80 (x$1000)