NIÑA MAGDALENE L.

ASENCI
Learning Objectives

1. Demonstrate the importance of forecasting for

business operations
2.Understand techniques to foresee the future.
3. Identify the major factors to consider when choosing a
forecasting technique.
4. Illustrate and distinguish between qualitative and
quantitative types of forecasting methods
 Sales will
- Process of predicting be $200
a future event Million
- Underlying basis of
all business decisions
Production
Inventory
Human resources
Finance

- Reduces uncertainty
to develop more
meaningful plans
2 Kinds of information when making
forecasts:

1.Current factors or conditions

2. Past experience in a similar situation.

 Economic forecasts - address
business cycle

Types of Technological forecasts – predict

technological change and new
Forecasts product sales

Demand forecasts – predict

existing product sales
Short- range forecast
o Up to 1 year; usually < 3 months
o Job scheduling, worker assignments
Medium-range forecast
o 3 months to 3 years
o Sales & production planning, budgeting
Long-range forecast
o 3+ years
o New product planning, facility location
 Medium/ Long range forecasts deal with
more comprehensive issues and support
management decisions regarding planning,
products and processes

 Short-term forecasts tend to be more

accurate than longer-term forecasts.
ELEMENTS OF A GOOD FORECAST:
1.It should be timely.
2.It should be accurate, and the degree of
accuracy should be stated.
3.It should be reliable.
4.It should be expressed in meaningful units
5.It should be in writing.
6.The forecasting technique should be simple
to understand and use.
7.It should be cost-effective.
1.Determine the purpose of the forecast.
2.Establish a time horizon.
3.Select a forecasting technique.
4.Obtain and analyze appropriate data.
5.Make the forecast.
6.Validate and implement results
FACTORS TO CONSIDER WHEN CHOOSING
FORECASTING TECHNIQUE:
a. Cost

b. Accuracy

c. Availability of historical data

d. Time needed to gather and analyze data

Qualitative Methods Quantitative Methods
Used when situation is  Used when situation is
vague and little data “stable” & historical
exist data exist
New products Existing products
New technology Current technology

Involves  Involves mathematical

intuition,personal techniques
experience, emotions
1. Executive Opinion
-pool opinions of high-level executives, sometimes
augment by statistical models

2. Salesforce Opinions
- good source of information because of their direct
contact with consumers
3. Consumer Surveys
-organizations use consumer survey to sample
consumer opinion

4. Delphi method
-series of questionnaires answered by
knowledgeable people
 Time series Associative model
1.Naive approach 1. Linear regression
2.Moving average

3.Exponential smoothing
What is a Time Series?
- sequence of measurements over time, usually obtained at
equally spaced intervals
 Daily ,monthly, quarterly, yearly

Example
Year: 1993 1994 1995 1996 1997
Sales:78.7 63.5 89.7 93.2 92.1

 The change in time series are affected by economic, social,

natural, industrial and political reasons.
Ft = a + bt, used to develop forecasts
when trend is present

Where
Ft = Forecast for period t
a = Value of Ft at t = 0
b = slope of the line
t = specified number of time periods from t=0
b= nΣty –ΣtΣy
nΣt2 – (Σt)2
a = Σy – bΣt
n

where
n= number of periods
y= value of time series
Example:

Cell phone sales for a Shanghai-based firm over the last 10

weeks are shown in the table. Determine the equation of
the trend line and predict sales for weeks 11 and 12.

Week Unit Sales b= nΣty –ΣtΣy

1 700 nΣt2 – (Σt)2
2 724
3 720
4 728
a = Σy – bΣt
5 740 n
6 742
7 758
8 750 Ft = a + bt
9 770
10 775
11 ?
12 ?
The variation of observations in a
time series which is unusual or
unexpected, unpredictable or
unforeseen events. It is also
termed as Irregular Variation.

Ex: Floods, fires, earthquakes,

revolutions, epidemics, strikes,
wars
1.Naive approach
-next value in a series will equal the previous value in
a comparable period

Example1: Data with stable series

 If demand for a product last week was 20 cases, the

forecast for this week is 20 cases.

 If May sales were 48, then June sales will be 48

 Example2: Data with trend

Period Actual Change from Forecast

Previous Value
t-2 50
t-1 53
(next)t +3 53+3= 56
2.Moving average
-based on an average of recent values and
used often for smoothing

Ft =MAn = A t -n + . . . + A t – 2 + A t – 1
n
Where

Ft = Forecast for time period t

MAn = n period moving average
A t – 1= Actual value in period t – 1
n =Number of periods (data points) in the
moving average
Compute a three-period moving average forecast given demand
for shopping carts for the last five periods.

Period Demand
1 84
2 80
3 80
4 85
5 81

F6 = 80 +85+ 81 = 82 F7 = ?
3
Disadvantages of Moving
Average Method
 Increasing n makes forecast less
sensitive to changes

 Require much historical data

WEIGHTED MOVING AVERAGE
a. Compute a weighted average forecast using a weight of
0.40 for the most recent period, 0.30 for the next most
recent, 0.20 for the next and 0.10 for the next.

Period Demand
1 84
2 80
3 86
4 80
5 82

F6 = 0.4(82) + 0.3(80) + 0.2(86) + 0.1(80) = 82

3.Exponential smoothing
-weighted average method based on previous
forecast plus a percentage of the forecast error

Next forecast = Previous forecast + α(Actual-(Previous forecast)

Example1:

The previous forecast was 84 units, actual

demand was 80 unit and α = 0.10. What would be
the new forecast?

 Ft = 84 + 0.10(80-84) = 83.6

What would be the new forecast of

actual demand was 76?
SIMPLE LINEAR REGRESSION
- simplest and most widely used form of
regression involves a linear relationship
between two variables

y = a + bx
SUMMARY OF FORECASTING TECHNIQUES:
Judgement/Opinion:
Consumer survey- Questioning consumers on future plans
Salesforce opinions- source of information because of their direct contact with consumers
Executive opinion- Finance, marketing and manufacturing managers join to
prepare forecast
Delphi technique- Series of questionnaires answered anonymously by
knowledgeable people

Statistical:
Time series:
Naïve - next value in a series will equal the previous value
Moving average - forecast is based on an average of recent values
Exponential smoothing - sophisticated form of weighted moving average

Associative model
Simple regression - values of one variable are used to predict values of a
dependent variable
ACCURACY AND CONTROL OF FORECASTS
Forecast error – difference between the value that
occurs and the value that was predicted for a given
time period. Error = Actual – Forecast

et = At – Ft
Example: If actual demand for a week is 200 units and the
forecast demand was 180 units. Compute for the forecast error.

If actual demand for a week is 30 units and the forecast

demand was 70 units. Compute for the forecast error.
MAD ( Mean Absolute Deviation)
- the average absolute forecast error

MAD =ΣIActualt – ForecasttI

n
Compute for MAD (Mean Absolute Deviation) for the following
data, showing actual and predicted numbers of accounts
serviced.
Period Actual Forecast Error IErrorI
(A-F)
1 217 215
2 213 216
3 216 215
4 210 214
5 213 211
6 219 214
7 216 217
8 212 216
Total
 Period Actual Forecast Error IErrorI
(A-F)
1 217 215 2 2
2 213 216 -3 3
3 216 215 1 1
4 210 214 -4 4
5 213 211 2 2
6 219 214 5 5
7 216 217 -1 1
8 212 216 -4 4
Total -2 22

MAD = 22 = 2.75
8
Technique Formula Definitions
Moving Average forecast Ft =MAn = A t -n + . . . + A t – 2 + A t – 1 n=number of periods
n A=demand in period

Exponential smoothing forecast Next forecast = P + α(A-P) α = smoothing factor

Linear trend forecast Ft = a + bt a=y intercept

b=slope
b= nΣty –ΣtΣy
nΣt2 – (Σt)2
a = Σy – bΣt
n

Linear regression forecast y= a + bx y=computed value of dependent

variable
b= nΣty –ΣtΣy x=independent variable
nΣt2 – (Σt)2 b=slope
a=value of y when x=0
a = Σy – bΣt
n

Forecast error Actual- Forecast

Given the following data:

Period Number of complaints

1 60
2 65
3 55
4 58
5 64

Prepare a forecast using each of these approaches:

A. Naive approach 64
B. Three-period moving average 59
C.Weighted average using weights of 0.5(most recent), 0.3 and 0.2 61
D. Exponential smoothing with a smoothing constant of 0.40. 61
 SALES DURING LAST YEAR
LAST YEAR REAL NEXT YEAR’S
SALES SEASONAL
FORECAST
Spring 200 220
Summer 350 385
Fall 300 330
Winter 150 165
Total Annual Sales 1,000 1,100
Annual increase of sales by 10%
What are the estimated seasonal sales amount for next year?