Ben Swanson owner and manager of Swansons Department Store h

Ben Swanson, owner and manager of Swanson’s Department Store, has decided to use statistical forecasting to get a better handle on the demand for his major products. However, Ben now needs to decide which forecasting method is most appropriate for each category of product. One category is major household appliances, such as washing machines, which have a relatively stable sales level. Monthly sales of washing machines last year are shown below.

January 23 February 24 March 22

April 28   May 22   June 27

July 20 August 26   September 21

October 29 November 23 December 28

T (b) Use the last-value method retrospectively to determine what the forecasts would have been for the last 11 months of last year. What is MAD?

T (c) Use the averaging method retrospectively to determine what the forecasts would have been for the last 11 months of last year. What is MAD?

T (d) Use the moving-average method with n 3 retrospectively to determine what the forecasts would have been for the last 9 months of last year. What is MAD?

(e) Use their MAD values to compare the three methods.

(f) Use their MSE values to compare the three methods.

(g) Do you feel comfortable in drawing a definitive conclusion about which of the three forecasting methods should be the most accurate in the future based on these 12 months of data?

I solved a, b, c & d I just need help comparing the MAD (e) and MSE (f) for the three following methods and solving (g).

b. Last value method retrospectively - MAD = 5.18       MSE = 30.64

c. Averaging method retrospectively - MAD 3              MSE = 11

d.Moving average with n=3 retrospectively - MAD 3.9           MSE = 17.41

Solution

(e) The higher the MAD values are the more chances of error are there. Since in averaging method retrospectively the numerical value of MAD is minimum, this case gives the least amount of error and thus among the given methods the averaging method most accurately represents the data set.

(f) In case of MSE the chances of fit are similar to that of MAD. The lower the MSE value the better the chances of method accurately representing the data set. Thus the order of methods from good fit to worst fit based on data are,

Average method retrospectively > Moving average with n=3 retrospectively > Last value method retrospectively.

(g)

According to MAD and MSE the Averaging value method is the most accurate method.