The accompanying data represent the yearly amount of solar power installed​ (in megawatts) in a particular area from 2000 through 2008. The trend forecasting equations below were​ found, where X is the number of years after 2000. Complete parts​ (a) through​ (d) below.

ModifyingAbove Upper Y with caret Subscript i Baseline equals negative 7.044 plus 28.2333 Upper X Subscript iYi=−7.044+28.2333Xi

ModifyingAbove Upper Y with caret Subscript i Baseline equals 25.76 plus 0.112 Upper X Subscript i Baseline plus 3.5152 Upper X Subscript i Superscript 2Yi=25.76+0.112Xi+3.5152X2i The accompanying data represent the yearly amount of solar power installed​ (in megawatts) in a particular area from 2000 through 2008. The trend forecasting equations below were​ found, where X is the number of years after 2000. Complete parts​ (a) through​ (d) below.

ModifyingAbove Upper Y with caret Subscript i Baseline equals negative 7.044 plus 28.2333 Upper X Subscript iYi=−7.044+28.2333Xi

ModifyingAbove Upper Y with caret Subscript i Baseline equals 25.76 plus 0.112 Upper X Subscript i Baseline plus 3.5152 Upper X Subscript i Superscript 2Yi=25.76+0.112Xi+3.5152X2i

Year

Amount

2000

23

2001

26

2002

46

2003

67

2004

84

2005

103

2006

144

2007

208

2008

252 Compute the standard error of the estimate

​(Upper S Subscript YXSYX​)

for each model.

Linear

Quadratic

Upper S Subscript YXSYX

enter your response here

enter your response here

​(Round to three decimal places as​ needed.)

Bold b.b.

Compute the

MADMAD

for each model.

Linear

Quadratic

MADMAD

enter your response here

enter your response here

​(Round to three decimal places as​ needed.)

Bold c.c.

On the basis of​ (a) and​ (b) and the principle of​ parsimony, which forecasting model would you​ select?

The model with the

▼   largest

smallest

values of

Upper S Subscript YXSYX

and

MADMAD

should be​ used, which is the

▼   linear

autoregressive

exponential

quadratic

model.