1234567 Super Bike Company (SBC) is a whole sale distributor of bicycles. Its distribution center in the town serves many retail stores within a 200-mile radius of the town. The data of bicycles ordered by the stores each month (i.e., monthly bicycle demand) during the last 5 years (year 1 year 5) is shown in Table 1 below. The analyst, Harry thought there may be a pattern of seasonal variation within the year, and decided to use the multiplicative seasonal variation method to forecast monthly demands of bicycles for next two years, year 6 and year 7. Harry started with calculating the aggregate demands, the annual demands for each of the past 5 years in order to forecast the annual demands for BOTH year 6 and year 7. Several approaches were considered for this purpose: (a) Naive approach (b) 3-year moving average 10 (c) Trend projection (d) Exponetial smoothing with trend, where Year 1's forecast = 12,000 units, Year 1's trend = 0, smoothing constant for forecast average = 0.6, and smoothing constant for trend = 0.3 12 (e) Associate method (i.e. simple linear regression) by relating bicycle annual demands to ad expense at the regional retail stores each year (see Tables 2.1 and 2.1) 13 14 Answer the following questions, Q1 ~ Q3. 15 16 16 17 18 Table 1 Q1: Select the applicable approach(es) from (a)~(e) that are able to meet Harry's purpose. Present the forecasted annual demads of year 6 and year 7 using the approach(es) you select. Table 2.1 19 Bicycles demands Ad expense (in $000) 20 21 Month 1 1244 Year 1 49.3 Month 2 878 Year 2 32.6 Month 3 1053 Year 3 41.6 Month 4 1080 Year 4 50.1 24 Year 1 Month 5 1541 Year 5 44.0 Month 6 871 Month 1001 Month 8 1265 Table 2.2 28 Month 9 879 Ad expense expected during 29 Month 10 1147 the year (in $000). 30 Month 11 988 Year 6 31 Month 12 1355 Year 7 45.0 48.1 32 Month 1 1200 33 Month 2 1046 1046 34 Month. 1025 1025 971 Month 4 8/1 26 Year 2 Month 5 Month 5 11/3 Month 6 1260 Month 6 1269 Month 7 909 Month 7 808 Month 8 Month 8 1396 1396 Month 9 Month 9 1030 1030 Month 10 962 963 42 Month 11 899 12 Month 12 1034 44 Month 1 1555 1339 45 Month 2 1164 Month 3 991 Month 4 1009 Year 3 Month 5 1148 Month 1037 Month 986 Month 8 1453 Month 9 1093 Month 10 1003 Month 11 Month 11 922 Month 12 Month 12 1406 Month Month 1 1541 1941 Month 2 1058 Month 3 1074 Month 4 959 Year 4 Month 5 1526 Month 6 1052 Month 7 903 Month 1127 Month 9 1057 Month 10 1086 Month 11 999 Month 12 1243 Month 1 1342 Month 2 1225 1225 Month 3 1015 1015 Month 4 1007 1007 Year 5 Month Month S 5 1562 1562 Month 6 1109 1105 Month Month 884 Month 1111 Month 9 1081 Month 10 907 78 Month 11 1067 79 Month 12 1143 80 81 Q2: For each approach you selected in Q1, calculate MAD to show its overall forecast error. Q3: Apparently, you will use the forecasts in Q1 from the approach with the least overall fortecast error found in Q2, to further calculate the monthly demand forecasts for year 6 and year 7, respectively. Show your calculation and result below.