Data on electric power consumption in a Midwestern town (in billions of kilowatt hours), income (in millions of dollars), and electricity prices (in cents per kilowatt hour) for the period 1987-2001 are shown below:
|
Year
|
Consumption
|
Income
|
Price
|
|
1987
|
407.9
|
944.0
|
2.09
|
|
1988
|
447.8
|
992.7
|
2.10
|
|
1989
|
479.1
|
1,077.6
|
2.19
|
|
1990
|
511.4
|
1,185.9
|
2.29
|
|
1991
|
554.2
|
1,326.4
|
2.38
|
|
1992
|
555.0
|
1,434.2
|
2.83
|
|
1993
|
586.1
|
1,594.2
|
3.21
|
|
1994
|
613.1
|
1,718.0
|
3.45
|
|
1995
|
652.3
|
1,918.3
|
3.78
|
|
1996
|
679.2
|
2,163.9
|
4.03
|
|
1997
|
696.0
|
2,417.8
|
4.43
|
|
1998
|
734.4
|
2,613.7
|
5.12
|
|
1999
|
730.5
|
2,957.8
|
5.80
|
|
2000
|
732.7
|
3,069.3
|
6.44
|
|
2001
|
750.9
|
3,304.8
|
6.83
|
a) Using regression analysis, estimate consumption as a linear function of income, price, and the previous year's consumption (assume consumption in 1986 was 367.7 Billion kilowatt hours). Write the equation, the t-stats, the R2, the Standard Error of the Estimate, and the F-statistic. Provide interpretation of the estimation (i.e., are the signs what you'd expect; what level of significance do the coefficients have, etc...). Provide a copy of your output from your regression analysis.
b) Assume that income in 2002 is $3,661.3 million and the price of electricity is 7.16 cent per kilowatt hour. Predict the consumption of electricity. How confident are you that the prediction is accurate?