Use the data from this problem to perform the following (note: your Weeks are in all one column 1 to 12, then second column is Passenger Miles for all 12 time periods) ie make Column 1 the week, Column 2 is the miles... you will create Column 3 forecast for #A, Column 4 Error for #A (so can figure MAD), Column 5 forecast for #B, Column 6 Error for #B, Column 7 forecast for #C, Column 8 Error for #C etc. Saves you from having to repeat the provided data multiple times (unless you wish to separate them)

A) Using a weighted moving average with 5 time periods, forecast the miles for week 13. Use a weight of 3 for the most recent, a weight of 3 for the next most recent, a weight of 2 for the next most, a weight of 1 for the next most, and a weight of 1 for the next ie 3, 3, 2, 1, 1 respectively.

B) Assuming an initial forecast for week 1 of 17,000 miles, use exponential smoothing to forecast miles for week 13. Use alpha = .2.

C) Assuming an initial forecast for week 1 of 17,000 miles, use exponential smoothing to forecast miles for week 13. Use alpha = .6.

D) Calculate the MAD for each forecasting technique and determine which is the most accurate? W

	actual passengers
Week	miles 1000s
1	17
2	21
3	19
4	23
5	18
6	16
7	20
8	18
9	22
10	20
11	15
12	22