FORECASTING MODEL

Definition

 Forecasting is a technique that uses historical data as inputs to make informed estimates
that are predictive in determining the direction of future trends.
 Businesses utilize forecasting to determine how to allocate their budgets or plan for
anticipated expenses for an upcoming period of time. This is typically based on the
projected demand for the goods and services offered.

3 METHODS OF FORECASTING MODEL

1. Moving Average Model

2. Exponential Smoothing
3. Trend Projection

 MOVING AVERAGE MODEL

- It is one of the simplest and easy-to-follow forecasting methods.
- Moving averages is a smoothing technique that looks at the underlying pattern of a set of
data to establish an estimate of future values. The most common types are the 3-month and
5-month moving averages.

EXAMPLE # 1:

Myles, the marketing manager of Silver Swan Soy Sauce, wants to forecast the sales of the
soy sauce on a monthly basis. He was able to monitor the monthly sales and is planning to use the
immediate past three months as the basis for his projection.

The Table below shows the monthly sales of the Silver Swan soy sauce:
 MONTHLY SALES
MONTH (This Year) SALES
January $5,000
February $8,000
March $7,000
April $8,000
May $6,000
June $9,000
July $7,000
August $9,000
September $9,000
October $8,000
November $10,000
December $8,000
Previous Months
Number 3

STEPS: (Moving Average Model)

1. Determine the Sales of the immediate past 3 months (to solve for the Previous Months)
Sales from:
 January to March= $5,000+$8,000+$7,000 = $20,000
 February to April= $8,000+$7,000+$8,000 = $23,000
…..
 September to November= $8,000+$10,000+$7,000 = $25,000
2. Determine the Sales Forecast (average)
Sales Forecast(average)= Previous Months’ Sales/3
 April- $20,000/3= $6,666.67
 May- $23,000/3= $7,666.67
…..
 December- $25,000/3= $8,333.33
 SALES FORECAST

MONTH (This Year) SALES PREVIOUS MONTHS SALES FORECAST

January $5,000
February $8,000 NO DATA
March $7,000
April $8,000 $20,000 $6,666.67
May $6,000 $23,000 $7,666.67
June $9,000 $21,000 $7,000.00
July $7,000 $23,000 $7,666.67
August $9,000 $22,000 $7,333.33
September $9,000 $25,000 $8,333.33
October $8,000 $25,000 $8,333.33
November $10,000 $26,000 $8,666.67
December $8,000 $27,000 $9,000.00

Thus, $9,000 sales forecast for December this year.

 EXPONENTIAL SMOOTHING MODEL

- It is a moving average technique that requires a minimum amount of past data.

EXAMPLE # 2:

Myles, the Marketing Manager of Silver Swan Soy Sauce in example # 1, wants to
include the past data forecasts in the new forecasts. He thinks that the smoothing constant is
75%. What is the forecast on monthly sales?

MONTHLY SALES
MONTH (This Year) SALES
January $5,000
February $8,000
March $7,000
April $8,000
May $6,000
June $9,000
July $7,000
August $9,000
September $9,000
 October $8,000
November $10,000
December $8,000
Smoothing Constant
a .75

STEPS: (Exponential Smoothing)

1. Compute for the Fraction of Error (a=.75) and Sales Forecast this month

Fraction of Error= a (Sales Last Month- Forecasts Last Month)

Sales Forecast this Month= Forecast last Month + Fraction of Error this Month

For the Month

of FRACTION OF ERROR SALES FORECAST THIS MONTH
January .75*(0-0)= 0 5+0= 5
February .75*(5-5)= 0 5+0= 5
March .75*(7-8)= 2.25 5+2.25= 7.25
April .75*(7-7.25)= -0.19 (-0.19)+7.25= 7.06
May .75*(8-7.06)= 0.71 7.06+0.70= 7.77

…... …... …...

December .75*(10-8.22)= 1.33 8.22+1.33= 9.56

SALES FORECAST
SMOOTHING 0.75
MONTH (This Year) SALES FRACTION OF ERROR SALES FORECAST
January $5 0 $5.00

February $8 0 $5.00

March $7 $2.25 $7.25

April $8 -$0.19 $7.06
May $6 $0.71 $7.77
June $9 -$1.32 $6.44
July $7 $1.92 $8.36
August $9 -$1.02 $7.34
September $9 $1.24 $8.59
 October $8 $0.31 $8.90
November $10 -$0.67 $8.22
December $8 $1.33 $9.56

Thus, the monthly Forecast Sales are:

January $5,000 July $8,360

February $5,000 August $7,340
March $7,250 September $8,590
April $7,060 October $8,900
May $7,770 November $8,220
June $6,440 December $9,560

 TREND PROJECTION MODEL

- It is a linear forecasting technique with time as the independent variable

EXAMPLE # 3:

Myles, the Marketing Manager of Silver Swan Soy Sauce in Examples #1 and #2, wants
to forecast the monthly sales for the next year. He thinks that there is a linear trend of sales over
time. Figures are summarized below:

MONTHLY SALES
MONTH (This Year) SALES ($000)
1 5000
2 8000
3 7000
4 8000
5 6000
6 9000
7 7000
8 9000
9 9000
10 8000
11 10000
12 8000
 Should Myles use the trend projection technique? What is the forecast on the monthly sales next
year?

SILVER SWAN SOY SAUCE

12

10

0
0 2 4 6 8 10 12 14 16

STEPS: (Trend Projection Model)

Trend line Formula:

𝑦 = 𝑎𝑥 + 𝑏

Find the value of the variables for the Trend Line Formula (𝑦 = 𝑎𝑥 + 𝑏)

1. Compute for xy, 𝑥 2 , 𝑦 2 by substituting Months this year as for the variable x and Sales
for the variable y
x- Month this Year
y- Sales
2. Find the mean/average of x and y
x= (Total of Months this year/No. of Months)
y= (Total Sales/No. of Months)
n=12(no. of months)
3. Compute for the coefficient a

𝒏(∑𝒙𝒚)−(∑𝒙∗∑𝒚)
𝒂= 𝒏(∑𝒙𝟐 )−(∑𝒙)𝟐

4. Solve for the coefficient b

 TREND LINE FORMULA

MONTHS THIS SALES

MONTHS YEAR (x) (y) 𝒙𝒚 𝒙𝟐 𝒚𝟐
January 1 5 5 1 25
February 2 8 16 4 64
March 3 7 21 9 49
April 4 8 32 16 64
May 5 6 30 25 36
June 6 9 54 36 81
July 7 7 49 49 49
August 8 9 72 64 81
September 9 9 81 81 81
October 10 8 80 100 64
November 11 10 110 121 100
December 12 8 96 144 64
TOTAL 78 94 646 650 758
MEAN/AVERAGE n COEFFICIENT
x 6.5 a 0.25
12
y 7.83 b 6.21

5. Solve the Sales Forecast for the next year using the Trend formula:
𝑦 = 𝑎𝑥 + 𝑏
January = 0.25(13) + 6.21= 9.46
February = 0.25(14) + 6.21= 9.71
March = 0.25(15) + 6.21 = 9.96
…..
December = 0.25(24) + 6.21= 12.21
Thus, the Sales Forecast for next year are:

MONTHS THIS YEAR SALES

MONTHS (x) (y)($000)
January 13 $9.46
February 14 $9.71
March 15 $9.96
April 16 $10.21
May 17 $10.46
June 18 $10.71
July 19 $10.96
August 20 $11.21
September 21 $11.46
October 22 $11.71
November 23 $11.96
December 24 $12.21
EXERCISE:

Henry, the Marketing Manager of Kojic Soap, wants to forecast the sales of the soap on
monthly basis. He was able to monitor the monthly sales and is planning to use the immediate
past three months as the basis for his projection.

Monthly Sales of the Kojic Soap are shown below:

MONTHLY SALES
MONTH (This Year) SALES
January $6,000
February $8,000
March $5,000
April $9,000
May $11,000
June $7,000
July $9,000
August $10,000
September $12,000
October $8,000
November $11,000
December $7,000

Required:

1. What is the Monthly Sales Forecast using Moving Average Model?

2. Using the same data, Henry thinks that the smoothing constant is 80%. Determine the
forecast on the monthly sales using the Exponential Smoothing Method.
3. Using the Trend Projection technique, what is the Monthly sales forecast for the next
year?
ANSWER:
1. Moving Average Model

MONTHLY SALES
MONTH (This Year) SALES
January $6,000
February $8,000
March $5,000
April $9,000
May $11,000
June $7,000
July $9,000
August $11,000
September $12,000
October $8,000
November $10,000
December $7,000
Previous Months
Number 3

SALES FORECAST
MONTH (This Year) SALES PREVIOUS MONTHS SALES FORECAST
January $6,000
February $8,000 NO DATA
March $5,000
April $9,000 $19,000 $6,333.33
May $11,000 $22,000 $7,333.33
June $7,000 $25,000 $8,333.33
July $9,000 $27,000 $9,000.00
August $11,000 $27,000 $9,000.00
September $12,000 $27,000 $9,000.00
October $8,000 $32,000 $10,666.67
November $10,000 $31,000 $10,333.33
December $7,000 $30,000 $10,000.00
SOLUTION:

Previous Months Sales Forecast

April- 6000 + 8000 + 5000 = $19000 19000/3 = 6333.33
May- 8000 + 5000 + 9000 = $22500 22000/3 = 7333.33
June- 5000 +9000 + 11000 = $25000 25000/3 = 8333.33
July- 9000 + 11000 + 7000 = $27000 27000/3 = 9000
August- 11000 + 7000 + 9000 = $27000 27000/3 = 9000
September- 7000 + 9000 + 11000 = $27000 27000/3 = 9000
October- 9000 + 11000 + 12000 = $32000 32000/3 = 10666.67
November- 11000 + 12000 + 8000 = $31000 31000/3 = 10333.33
December- 12000 + 8000+ 10000 = $30000 30000/3 = 10000

2. Exponential Smoothing

SALES FORECAST
MONTHLY SALES 0.8
MONTH (This Year) SALES($000) FRACTION OF ERROR SALES FORECAST
1 $6 0 $6.00
2 $8 0 $6.00
3 $5 $1.60 $7.60
4 $9 -$2.08 $5.52
5 $11 $2.78 $8.30
6 $7 $2.16 $10.46
7 $9 -$2.77 $7.69
8 $11 $1.05 $8.74
9 $12 $1.81 $10.55
10 $8 $1.16 $11.71
11 $10 -$2.97 8.74
12 $7 $1.01 $$9.75