Download as docx, pdf, or txt
Download as docx, pdf, or txt
You are on page 1of 3

QUANTITATIVE FORECASTING TECHNIQUES

1. MOVING AVERAGE
-is a calculation to analyze data points by creating a series of averages of different
subsets of the full data set
-is a technique to get an overall idea of the trends in a data set

formula:

∑ 𝑑𝑒𝑚𝑎𝑛𝑑 𝑖𝑛 𝑝𝑟𝑒𝑣𝑖𝑜𝑢𝑠 𝑛 𝑝𝑒𝑟𝑖𝑜𝑑𝑠


MA = 𝑛(𝑛𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑝𝑒𝑟𝑖𝑜𝑑𝑠 𝑖𝑛 𝑡ℎ𝑒 𝑚𝑜𝑣𝑖𝑛𝑔 𝑎𝑣𝑒𝑟𝑎𝑔𝑒)
Example:

Period Actual 3-period 4-period


demand MA MA
1 42
2 40
3 43
4 40 MA=
42+40+43
= 41.67
3
5 41 MA=
40+43+40
= 41 MA=
42+40+43+40
= 41.25
3 4

2. WEIGHTED MOVING AVERAGE

formula:

∑(𝑤𝑒𝑖𝑔ℎ𝑡 𝑓𝑜𝑟 𝑝𝑒𝑟𝑖𝑜𝑑 𝑛)𝑥(𝑑𝑒𝑚𝑎𝑛𝑑 𝑖𝑛 𝑝𝑒𝑟𝑖𝑜𝑑 𝑛)


WMA = ∑ 𝑤𝑒𝑖𝑔ℎ𝑡𝑠

Example:

Period Actual 3-period


demand WMA
1 42
2 40
3 43
4 40 WMA=
42(1)+40(2)+43(3)
= 41.83
6
5 41 MA=
40(1)+43(2)+40(3)
= 41
6

3. EXPONENTIAL SMOOTHING
-is a time series forecasting method for univariate data that can be extended to
support data with a systematic trend or seasonal component

formula:

Ft = 𝐹𝑡−1 +∝ (𝐴𝑡−1 − 𝐹𝑡−1 )


Where: Ft = the new forecast
Ft-1 = the previous forecast
∝ = (alpha) smoothing constant
At-1 = previous period’s actual data

Example:
Period Actual demand Forecast Error
∝= 10%
1 42 0 0
2 40 42 -2
3 43 41.8 1.2
4 40 41.92 -1.92
5 41 41.73 -0.73

4. TIME SERIES USING TREND PROJECTION MODEL AND REGRESSION ANALYSIS


-is a series of data points listed or graphed in time order

TREND LINE FORMULA:

𝑦 = 𝑎 + 𝑏𝑥

where: y = quantity being forecast


𝑎 = the point at which the trend line intercepts the vertical axis
b = rate of change or slope of the line
x = the independent variable, usually represents time

to compute for b in the trend line:

∑ 𝑥𝑦 − 𝑛𝑥𝑦
̅̅̅
𝑏=
∑ 𝑥 − 𝑛𝑥̅ 2
2

where: x = time
n = number of items
∑𝑥
x = mean of x 𝑥̅ = 𝑛

∑𝑦
y = mean of y 𝑦̅ =
𝑛

to compute for 𝑎 in the trend line:

𝑎 = ̅𝑦 − 𝑏𝑥̅
Example:

Suppose that the sales of ABC Hypermart in the past 6 months were recorded as follows:

Month Actual sales x xy x2


(y)
May 250 0 0 0
June 300 1 300 1
July 275 2 550 4
August 325 3 975 9
September 290 4 1160 16
October 350 5 1750 15
∑ 𝑦 = 1790 ∑ 𝑥 = 15 ∑ 𝑥𝑦 = 4735 ∑ 𝑥 2 = 55

15 1790
n=6 ̅𝑥 = 6
= 2.5 𝑦̅ = 6
= 298.33

4735−[6(2.5)(298.33)]
𝑏= 55−[6(2.5)2 ]
= 14.86

𝑎 = 298.33 − [14.86(2.5)] = 261.18

therefore, the trend line equation is : 𝑦 = 𝑎 + 𝑏𝑥


= 261.18 + 14.86 𝑥

we can now forecast the value of y for any value of x (the time)

a.) suppose we wanted to forecast the sale of ABC Hypermart for the month of
November

Sol’n: From May to November is 6 months, hence x = 6


𝑦 = 261.18 + 14.86 𝑥
= 261.18 + 14.86 (6)
= 350.34

b.) forecast the sale in January of the following year

Sol’n: From May to January of the following year, x = 8


y = 261.18 + 14.86 x
= 261.18 + 14.86 (8)
= 380.06

You might also like