Set up an optimization model in Excel and use the solver to match the SAS results given in gigure 6.36.

Year Quarter 1 1 344.39 (= y) 2 313.45 (= ys) 3 246.84 4 277.01 5 365.10 6 267.00 7 467.06 8 336.56 9 440.00 10 434.66 (= y37) Quarter 2 246.63 (= y2) 189.76 (= y) 209.00 197.98 207.51 230.28 306.03 196.67 315.04 399.66 (= y38) Quarter 3 131.53 (= y3) 179.10 (= y) 51.21 50.68 54.63 230.32 253.23 152.15 216.42 330.80 (= y39) Quarter 4 288.87 (= ya) 221.10 = ys) 133.89 218.08 214.09 426.41 279.46 319.67 339.78 539.78 (= yao) FIGURE 6.36 SAS PROC ARIMA analysis of the energy bill data using the model Ye= Bo + B.t +B_t? +BQ, + B.Q2 + BsQ3 + , where e = 018-1 + a Paraneter Estimate Lag Variable MU AR1,1 NUMI NUM2 NUM3 NUM4 NUMS 283.94906 0.59408 -9.21968 0.35348 70.10688 -35.42856 -126.52509 Conditional Least Squares Estination Standard Approx Error t Value Pr > It: 47.55019 5.97 <.0001 o-ooooo y tine times q1 q2 q3 obs forecast constant estimate variance std error confidence limits year quarter ys y37 y2 y38 y3 y39 ya yao figure sas proc arima analysis of the energy bill data using model ye="Bo" b.t b.q2 bsq3 where e="018-1" a paraneter lag variable mu ar1 numi num2 num3 num4 nums conditional least squares estination standard approx t value pr> It: 47.55019 5.97 <.0001 o-ooooo y tine times q1 q2 q3 obs forecast constant estimate variance std error confidence limits>