Management Science

Chapter 4: Forecasting

Fershie D. Yap, Ph.D.

College of Accountancy and Business Administration
 Forecasting

An operational technique used as a basis for

management planning and decision making
It is an art and science of predicting future events
It involves historical data in projecting the future
It is a mathematical model adjusted by
managers’ good judgment
 Forecasting:
Features

It is concerned with future events

It is necessary for planning process
The impact of future events has to be considered
in the planning process
It is a guessing of future events
It considers all the factors which affect
organizational functions
 Forecasting:
Applicability is Dependent on

1. Time frame of the forecast

Short range – immediate future, daily
operations
Medium range – from 1 or 2 mos to 1 year
Long range – longer than or 2 years (plan
new products for changing markets, build new
facilities, or secure long term financing
 Forecasting:
Applicability is Dependent on
2. The existence of patterns in the forecast
Trend – long term movement of the item
forecasts
Random variations – unpredictable movements
Cycle – undulating movement in demand (more
than 1 year)
Seasonal pattern – hesitating movement in
demand (short run)
3. The number of variables
 Forecasting:
Applicability is Dependent on

Trend Cycle

Seasonal Trend with

Pattern seasonal
pattern
 Forecasting:
Process

1. Thorough preparation of foundation

2. Estimation of future – consultation and
communication with key personnel
3. Collection of results
4. Comparison of results – actual results versus
estimated results
5. Refining the forecast – through deviations
 Forecasting: Importance
1. Pivotal role in an organization – planning is based on
forecasting
2. Development of a business - performance of specified
objectives depends upon the proper forecasting
3. Coordination – collection of information (internal and
external factors)
4. Effective control – can ascertain strength and weaknesses
5. Key to success – provides clues and reduce risk
6. Implementation of project – entrepreneur gain experience
to succeed
7. Primary to planning
 Forecasting:
Strategic Importance
1. Human Resource – hiring, training, and laying off
workers all depends on anticipated demand
2. Capacity – inadequacy results to shortages and
undependable delivery, loss of customers and
market share
3. Supply Chain Management – good supplier
relations and price advantages
 Forecasting:
Types
1. Economic forecasts – addresses the business
cycle by predicting inflation rates, money
suppliers, housing starts, and other planning
indicators
2. Technological forecasts – concerned with rates of
technological progress (new products)
3. Demand forecasts – projections of demand for a
company’s products or services (sales forecast,
production, capacity, scheduling systems)
 Forecasting:
Methods
1. Time Series – statistical techniques that uses
historical data to predict future behavior with
time as factor (short and long range forecasting)
2. Regression (causal) methods – attempt to
develop a mathematical relationship between
the item being forecast and factors that cause it
3. Qualitative methods – use management
judgment, expertise, and opinion (the jury of
executive opinion)
 Forecasting:
Time Series Methods (The Moving Average)

The Moving Average – uses historical actual data

values to develop a forecast; computed for specific
periods such as 3 or 5 months
1. Simple Moving Average
2. Weighted Moving Average
 Forecasting:
Time Series Methods (The Moving Average)
1. Simple Moving Average – computed for specified
time period
 Forecasting:
Time Series Methods (The Moving Average)
1. Simple Moving Average (ex. Delivery orders for 10-mo. period)
 Forecasting:
Time Series Methods (The Moving Average)
2. Weighted moving average - puts more weight on recent data
and less on past data.
 Forecasting:
Time Series Methods (The Moving Average)
2. Weighted moving average:
Example, if the Instant Paper Clip Supply Company wants to
compute a 3-month WMA with a weight of 50% for the
October data, a weight of 33% for the September data, and a
weight of 17% for August, compute the WMA 3-month
 Forecasting:
Time Series Methods (The Moving Average)
2. Weighted moving average:
 Forecasting:
Time Series Methods (Exponential Smoothing)

Exponential Smoothing
- an averaging method that weights the most recent
past data more strongly than more distant past data
- Two forms: simple exponential smoothing and
adjusted exponential smoothing
 Forecasting:
Time Series Methods (Exponential Smoothing)

Exponential Smoothing (Simple Exponential Smoothing)

α = 0 & 1 (0.01-0.50)

If α=0.20

Ft+1 = 0.20Dt + 0.80Ft

The higher α is, the closer α to 1, the more sensitive the forecast to recent demand
The closer α to zero, the > will be the smoothing effect
 Forecasting:
Time Series Methods (Exponential Smoothing)
Exponential Smoothing (Simple Exponential Smoothing)

February forecast
 Forecasting:
Time Series Methods (Exponential Smoothing)
Exponential Smoothing (Simple Exponential Smoothing)

Period 3 forecast
 Forecasting:
Time Series Methods (Exponential Smoothing)
Exponential Smoothing (Simple Exponential Smoothing)

Period 13 forecast
 Forecasting:
Time Series Methods (Exponential Smoothing)

Exponential Smoothing
(Simple Exponential
Smoothing)
 Forecasting:
Time Series Methods (Exponential Smoothing)
Exponential Smoothing
(Simple Exponential Smoothing)
 Forecasting:
Time Series Methods (Exponential Smoothing)
Exponential Smoothing (Adjusted Exponential Smoothing)
- With a trend adjustment factor added to it

Forecast Model for Trend

β=0&1
High β = trend changes more than low β
Closer β to 1 = the stronger a trend
 Forecasting:
Time Series Methods (Exponential Smoothing)
Exponential Smoothing
(Adjusted Exponential
Smoothing)
Example: PM Computer
Services now wants to
develop an adjusted
exponentially smoothed
forecast, using the same 12
months of demand.
 Forecasting:
Time Series Methods (Exponential Smoothing)
Exponential Smoothing
(Adjusted Exponential Smoothing)

-T3 = β (F3 – F2) + (1- β)T2

= (0.30)(38.50-37) + (0.70)(0)
= 0.45

-AF3 = F3 + T3 = 38.5 + 0.45

= 38.95 units
 Forecasting:
Time Series Methods (Exponential Smoothing)
Exponential Smoothing
(Adjusted Exponential Smoothing)

-T13 = β (F13 – F12) + (1- β)T12

= (0.30)(53.61-53.21) + (0.70)(1.77)
= 1.36

-AF13 = F13 + T13 = 53.61 + 1.36

= 54.96 units
 Forecasting:
Time Series Methods (Exponential Smoothing)
Exponential Smoothing
(Adjusted Exponential Smoothing)
 Forecasting:
Time Series Methods (Linear Trend Line)
Linear Regression
- causal method of forecasting in which a mathematical
relationship is developed between demand and some other
factor that causes demand behavior.
- when demand displays an obvious trend over time, a least
squares regression line or linear trend line, can be used to forecast
demand
- linear trend is a linear regression model that relates demand to
time
 Forecasting:
Time Series Methods (Linear Trend Line)
Linear Regression

Linear Equation Trend Line

 Forecasting:
Time Series Methods (Linear Trend Line)
Linear Regression

Trend Line
 Forecasting:
Time Series Methods (Linear Trend Line)
Linear Regression Trend Line
 Forecasting:
Time Series Methods (Linear Trend Line)
Linear Regression

y=35.2+1.72x Linear Trend Line

Forecast for
y=35.2+1.72(13)=57.56 Period 13
 Forecasting:
Time Series Methods (Linear Trend Line)
Linear Regression
 Forecasting:
Time Series Methods (Seasonal Adjustments)
Seasonal Adjustments
- Repetitive up-and-down movement in demand
- Adjusting seasonality by multiplying it by seasonal factor
- Resulting seasonal factors = 0 & 1

Seasonal Factor

Seasonally Adjusted Factor = (Seasonal Factor)(Demand Forecast)

Or
SF = S x F
 Forecasting:
Time Series Methods (Seasonal Adjustments)
Seasonal Adjustments: Example

Seasonal Factor

Demand forecast for 2006 = 58.17

 Forecasting:
Time Series Methods (Seasonal Adjustments)
Seasonal Adjustments: Example SF = S x F

Demand forecast
for 2006 = 58.17
 Forecasting:
Forecast Accuracy (Forecast Error)
Forecast Error - the difference between the forecast and actual
demand
Measures of Forecast Error:
1. Mean Absolute Deviation (MAD)
2. Mean Absolute Percent Deviation (MAPD)
3. Cumulative Error (E)
4. Average Error or Bias (Ē)
5. Mean Squared Error (MSE)
 Forecasting:
Forecast Accuracy(Forecast Error)
1. Mean Absolute Deviation (MAD) – the average, absolute
difference between the forecast and the demand
 Forecasting:
Forecast Accuracy(Forecast Error)
2. Mean Absolute Percent Deviation (MAPD)
– the absolute error as a percentage of
demand

3. Cumulative Error (E)

– sum of the forecast error

4. Average Error or Bias (Ē)- per-period

average of cumulative error
 Forecasting:
Forecast Accuracy(Forecast Error)
5. Mean Squared Error (MSE)
- each individual error
value is squared, summed
and averaged
(the smaller MSE, the better)
 Forecasting:
Forecast Accuracy(Forecast Error)
5. Mean Squared Error (MSE)
 Forecasting:
Regression Methods Linear Equation
Regression
- is a forecasting technique
that measures the relationship
of one variable to one or more
other variables
Simple Linear Regression
- simplest form
- relates one dependent
variable to one independent
variable in the form of a
linear equation
 Forecasting:
Regression Methods
Simple Linear Regression
 Forecasting:
Regression Methods
Simple Linear Regression

y = 18.46 + 4.06x linear equation line

x=7 (wins),
y = 18.46 + 4.06(7) forecast for
= 46.88 or 46,880 attendance
 Forecasting:
Regression Methods
Correlation – a measure of the strength of the relationship between
the independent and dependent variables

r = -1.00 and +1.00,

r = ±1.00 strong linear relationship between the variables
 Forecasting:
Regression Methods
Correlation – Example
 Forecasting:
Regression Methods
Correlation – Example
 Forecasting:
Regression Methods
Coefficient of Determination (r2)
- another measure of the strength of the relationship between
the variables in a linear regression equation
- percentage of the variation in the dependent variable that
results from independent variable
- the ratio of the explained variation to the total variation
 Forecasting:
Regression Methods
Coefficient of Determination
- another measure of the strength of the relationship between
the variables in a linear regression equation
- percentage of the variation in the dependent variable that
results from independent variable