Question

A convenience store recently started to carry a new brand of soft drink. Management is interested in estimating future sales volume to detemine whether it should continue another brand. The following table provides the number of cans sold per week. Use both the trend projection with regression and the exponential smoothing (let 0.4 with an initial forecast for week 1 of 577) methods to forecast demand for week 13. Compare these methods by using the mean absolute deviation and mean absolute percent error performance criteria. Does your analysis suggest that sales are trending and if so, by how much? carry the new brand or replace it with Period 1 2 3. 4 6 7 8 9 10 11 12 577 647 Sales 623 641 733 16 746 719 674 44 662 734 (i) Obtain the trend projection with regression forecast. The forecast for week 13 is 741. (Enter your response rounded to the nearest whole number.) Specify the mean absolute deviation (MAD) and mean absolute percent error (MAPE). (Enter your responses rounded to two decimal places.) MAD MAPE

Answer 1

i)

C D G H (Abs(error)/Sales)*100 Trend forecast error abs(error) Period(x) Sales(y) 1 x2 xy 621.5121 44.51 2 577 577 1 -44.5121 7.71 3 2 623 1246 4 631.479 -8.479 8.48 1.36 4 3 641 1923 641.4459 -0.4459 0.45 0.07 5 4 733 2932 16 651.4128 81.5872 81,59 11.13 5 647 3235 25 661.379714.3797 14.38 2.22 7 6 616 3696 36 671.3466 -55.3466 55.35 8.98 64.69 8 7 746 5222 49 681.3135 64.6865 8.67 719 5752 64 691.2804 27.7196 27.72 3.86 10 9 674 6066 81 701.2473 -27.2473 27.25 4.04 32.79 11 10 744 7440 100 711.2142 32.7858 4.41 721.1811 -59.1811 59.18 12 11 662 7282 121 8.94 13 12 734 8808 144 731.148 2.852 2,85 0.39 741.1149 14 13 15 Total 91 8116 54179 650 419.22 61.79 16 17 MAD 34.94 18 MAPE 5.15 19 20

MAD = Sum (abs(error))/n

MAPE = [abs(error)/sales]*100/n