In: Economics
Given these sales figures over the last 7 weeks, your boss needs you to test two different forecasting methods (parts a and b below) to determine which method is best. For your measure of “best”, recommend to your boss that the company should use the method with the lowest mean absolute deviation (MAD). Then use that method to provide your forecast for week 8 in part c.
Week Unit Sold
1 9,040
2 11,150
3 9,420
4 10,310
5 11,950
6 13,050
7 12,470
Calculate the MAD for the 4 period moving average forecasting technique.
Calculate the MAD for the 2 period weighted moving average forecasting technique. Use weights of 0.6 and 0.4, with the most recent observation weighted higher .
How many units would you forecast will be sold in week 8?
MAD is the technique used where we see the the deviation a week has from the mean. Zero deviation from the eman is the best observation.
First we will find the mean of the observations given to us.
(9040+11150+9420+10310+11950+13050+12470) / 7 = 11055.7
Now for part 1) we will use 4 period to forecast the units sold. For this we will take week 1 + week 2+ week 3+ week 4 and divide it by 4 to get the forecast for the next wek that is week 5.
The we will subtract the actual unit sold of week 5 and the forcasted week 5. And get the absolute value.
the below table shows the calculation
Fo the part 2, it is given that we should take just last two weeks to find the mean. In this the latest is given more weigtage that is .6 and the later one gets .4 weight.
So for week the forecast would be =0.6 * week 2 + 0.4* week1
and then the same we we subract and get the deviation.
Calculation is below.
When we compare teh deviation of both the part 1 & 2 we see that part 2 has alower deviation that part 1 so the company should follow the techhinque used in part 2 where more importance is goven to the latest data.
Fo week 8 the forcast should be= .6 *12470 +.4* 13050= 12702 (answer)