Question

A brokerage company is interested in forecasting the number of new accounts the office will obtain next month. It has collected the following data for the past 12 months.

Month	Accounts
1	19
2	20
3	21
4	25
5	26
6	24
7	24
8	21
9	27
10	30
11	24
12	30

1. Use a four month moving average starting with the month of May to forecast the number of new accounts the company may expect to receive next month (Year 2, Month 1). Compute MAD, MSE, MAPE, and MPE.
2. Use exponential smoothing with a smoothing constant of 0.15 and an initial value of 20 to forecast the number of new accounts the company may expect to receive next month. Compute MAD, MSE, MAPE, and MPE.
3. Which forecasting technique would provide the best results? Justify your answer.

Answer 1

Question: A brokerage company is interested in forecasting the number of new accounts the office will obtain next month. It has collected the following data for the past 12 months.

Answer: (Note: Since the moving average computes from May we take error calculations of the corresponding forecast that is Exponential Smoothing from May as well to balance the conclusion.)

Overall calculations:

Excel formulas used:

Next Year January Forecast:

4-Moving Average = 27.75

Exponential Smoothing = 24.81

Formulas:

Forecast:

Moving Average = (A1 + A2 + A3 + ... + An) / n

Here: A = Accounts Value

Exponential Smoothing = x At + (1 - ) x Ft

Here: At = Accounts Value; Ft = Forecasted Value

Error Computation:

Mean Absolute Deviation (MAD) = |Error| / n

Here: |Error| = |Accounts - Forecast|

Mean Squared Error (MSE) = Error^2

Here: Error^2 = (Accounts - Forecast)^2

Mean Percent Error (MPE) = Error%

Here: Error% = (Error / Accounts) x 100

Mean Absolute Percent Error (MAPE) = Error Abs %

Here: Error Abs % = (|Error| / Accounts) x 100

Conclusion: 4-Moving Average Forecast Model provides better forecasts in terms of MAD, MSE, MAPE, and MPE. Since MAD, MSE, MAPE, MPE is lowest and closest to the actual values that are "Accounts" values.

General rule: MAD, MSE, MAPE, MPS is the different forecasting error measure to compute the average distance between the actual value and their corresponding mean. The greater the mean value the greater variability in the computed data. Lowest MAD, MSE, MAPE, MPS values the lesser and more closest to the actual values.

(Note: Since the Moving Average Forecast computes from May we take error calculations of the corresponding forecast that is Exponential Smoothing from May as well to balance the conclusion.)