Chapter 1
Forecasting
1
Forecasting
1.1 Introduction
➢ Forecasting is the process of making predictions of the future
based on past and present data and most commonly by
analysis of trends.
2
Forecasting
1.1 Introduction
Forecasting is important in many aspects of our lives. As
individuals, we try to predict success in our marriages,
occupations وظائف, and investments. Organizations invest
enormous amounts based on forecasts for new products,
factories, and contracts with executives. Government agencies
need forecasts of the economy, environmental impacts, and
effects of proposed social programs.
3
Forecasting
1.1 Introduction
Forecasting is also used to predict profits, revenues, costs,
productivity changes, prices, resources, and economic
indicators ( such as inflations, interest rates, and etc.)
4
Forecasting
1.1 Introduction
Decision makers need forecasts only if there is uncertainty about
the future. Thus, we have no need to forecast whether the sun
will rise from the east. There is also no uncertainty when events
can be controlled; for example, you do not need to predict the
arrival time of employees to their work place. Many
decisions, however, involve uncertainty, and in these cases,
formal forecasting procedures can be useful.
5
Forecasting
1.1 Introduction
Forecasting is often confused with planning. Planning concerns
what the world should look like, while forecasting is about what
it will look like.
Forecasting, goals and planning
Forecasting is a common statistical task in business, where it
helps inform decisions about scheduling of production,
transportation and personnel, and provides a guide to long-term
strategic planning
6
Forecasting
1.1 Introduction
However, business forecasting is often done poorly and is
frequently confused with planning and goals. They are three
different things.
Forecasting is about predicting the future as accurately as
possible, given all the information available including historical
data and knowledge of any future events that might impact the
forecasts.
7
Forecasting
1.1 Introduction
Goals are what you would like to happen. Goals should be linked
to forecasts and plans, but this does not always occur. Too often,
goals are set without any plan for how to achieve them, and no
forecasts for whether they are realistic.
Planning is a response to forecasts and goals. Planning involves
determining the appropriate actions that are required to make
your forecasts match your goals.
8
Forecasting
1.1 Introduction
Forecasting should be an integral part of the decision-making
activities of management, as it can play an important role in
many areas of a company. Modern organizations require short-,
medium- and long-term forecasts, depending on the specific
application.
9
Forecasting
1.1 Introduction
Short-term forecasts are needed for scheduling of personnel,
production and transportation.
As part of the scheduling process, forecasts of demand are often
also required. It covers days or weeks
Medium-term forecasts are needed to determine future resource
requirements in order to purchase raw materials, hire personnel,
or buy machinery and equipment. It covers months
10
Forecasting
1.1 Introduction
Long-term forecasts are used in strategic planning. Such
decisions must take account of market opportunities,
environmental factors and internal resources. It covers a year or
more
11
Forecasting
1.2 Common forecasts features
1- Conditions existed in the past will continue to exist in the
future.
2- Always there is a difference between actual results and
predicted values.
3- Forecast accuracy usually decreases with increasing time
horizon covered by the forecast.
12
Forecasting
1.3 Elements of a good forecast
➢ Timely
➢ Accurate
➢ Reliable
➢ Expressed in meaningful units.
➢ In writing
➢ Simple to understand and use
➢ Cost effective
13
Forecasting
1.4 Steps in the forecasting process
1- Determine the purpose of the forecast.
2- Establish a time horizon.
3- Select a forecasting technique.
4- Gather and analyze relevant data.
5- Make the forecast.
6- Monitor the forecast.
7- Modify the forecast if needed.
14
Forecasting
1.5 Approaches to forecasting
1.5.1 Qualitative approach.
1.5.2 Quantitative approach.
15
Forecasting
1.5.1 Qualitative approach.
Qualitative forecasting techniques are subjective, based on the
opinion and judgment of consumers, experts; they are
appropriate when past data are not available. They are usually
applied to intermediate- or long-range decisions.
Qualitative forecasting techniques rely on analysis of subjective
inputs which often lack precise numerical description . It is
known as judgmental forecast.
16
Forecasting
Judgmental or opinion forecasts occur in some situations like:
❑ Quick forecast is required while no available quantitative data
❑ When political and economic conditions are changing and
available data may be obsolete.
❑ The introduction of new product or the redesign of existing
one
17
Forecasting
Who makes judgmental or opinion forecasts ?
Forecasts are based on executive opinions, customer surveys,
opinion of the sales staff, and opinions of experts.
➢ Executive opinions
A small group of upper-level managers meet and collectively
develop a forecast. This approach is used as part of long range
planning and new product development. There is a risk that
the view of one person may prevail.
18
Forecasting
➢ Sales force opinions
Good source of information due to direct contact with
consumers. Some drawbacks are
-may be unable to distinguish between what customers would
like to do and what actually will do
- May be influenced by recent experiences ( low sales or good
sales )
19
Forecasting
➢ Customer surveys
Collecting information from samples of customers directly
➢ Experts opinions
May include advice on political or economic conditions
20
Forecasting
1.5.2 Quantitative approach.
Rely on analysis of objective inputs which use numerical data
obtained from history, or time series . Quantitative forecasting
models are used to forecast future data as a function of past
data. They are appropriate to use when past numerical data is
available and when it is reasonable to assume that some of the
patterns in the data are expected to continue into the future.
These methods are usually applied to short- or intermediate-
range decisions.
21
Forecasting
Quantitative forecasts are based on time series data
➢ Time series data are time-ordered sequence of observations
taken at regular intervals over a period of time (hourly, daily,
weekly, …etc.)
➢ The data may be measurements of demand,, revenues, profits,
productivity, …etc.)
➢ The assumption is that future values can be estimated from
past values
22
Forecasting
Time series data can take different patterns such as:
➢ Trend… gradual long term movement ( upward or
downward)
➢ Seasonal variations…short term, fairly regular variations.
➢ Cyclic variations…wave-like variations more than one year’s
duration
➢ Irregular variation… due to unusual circumstances
➢ Random variation…due to common causes of variation
23
Forecasting
Time-Series Forecasting
Time-series forecasting is a quantitative forecasting technique. It
measures data gathered over time to identify trends. The data
may be taken over any interval: hourly; daily; weekly;
monthly; yearly; or longer.
Trend, cyclical, seasonal and irregular components make up the
time series. The trend component refers to the data's gradual
shifting over time. It is often shown as an upward- or
downward-sloping line to represent increasing or decreasing
trends, respectively.
24
Forecasting
Time-Series Forecasting
Cyclical components lie above or below the trend line and repeat
for a year or longer. The business cycle illustrates a cyclical
component. Seasonal components are similar to cyclicals in
their repetitive nature, but they occur in one-year periods. The
annual increase in gas prices during the summer driving
season and the corresponding decrease during the winter
months is an example of a seasonal event. Irregular
components happen randomly and cannot be predicted.
25
Forecasting
1.6 Techniques for forecasting
The selection of a forecasting technique depends on many factors:
➢ the context of the forecast,
➢ the relevance and availability of historical data,
➢ the degree of accuracy desirable,
➢ the time period to be forecast,
➢ the cost/ benefit (or value) of the forecast to the company, and
➢ the time available for making the analysis.
26
Forecasting
1.6 Techniques for forecasting
Successful forecasting begins with a collaboration between the
manager and the forecaster, in which they work out answers to the
following questions.
1. What is the purpose of the forecast—how is it to be used? This
determines the accuracy and power required of the techniques, and
hence governs selection. Deciding whether to enter a business may
require only a rather gross estimate of the size of the market,
whereas a forecast made for budgeting purposes should be quite
accurate. The appropriate techniques differ accordingly. 27
Forecasting
1.6 Techniques for forecasting
2. What are the dynamics and components of the system for which
the forecast will be made? This clarifies the relationships of
interacting variables. Generally, the manager and the forecaster
must review a flow chart that shows the relative positions of the
different elements of the distribution system, sales system,
production system, or whatever is being studied.
28
Forecasting
1.6 Techniques for forecasting
3. How important is the past in estimating the future? Significant
changes in the system—new products, new competitive strategies,
and so forth.
29
Forecasting
1.6 Techniques for forecasting
1- Naïve forecasts
2- Simple moving average
3- weighted moving average
4- Exponential smoothing
5- Trend forecasting
30
Forecasting
1.6.1 Naïve forecasts
Naïve forecast for any period equals the previous period’s actual value. It is the
simplest forecasting technique, quick and easy but not accurate
❑ If the data series show no trend, then: Ft = At-1
❑ If the data series show trend, then: Ft = At-1 + (At-1 - At-2)
Where Ft is the forecasted value for period t and At-1 and At-2 are, respectively,
the actual values for period's t-1 and t-2.
31
Forecasting
1.6.2 Simple Moving Averages
❑ It averages a number of recent actual values, updated as new
values become available. The simple moving average forecast
can be computed using the following equation:
Ft = MAn = ∑ Ai / n
it is called the n-points simple moving average
Where, i = “Age “of the data (I = 1, 2, 3…)
n = Number of periods in the moving average
Ai = Actual value with age i
Ft = Forecast for the time period t 32
Forecasting
1.6.2 Simple Moving Averages
❑ It averages a number of recent actual values, updated as new
values become available. The simple moving average forecast
can be computed using the following equation:
Ft = MAn = ∑ Ai / n
it is called the n-points simple moving average
Where, i = “Age “of the data (I = 1, 2, 3…)
n = Number of periods in the moving average
Ai = Actual value with age i
Ft = Forecast for the time period t 33
Forecasting
1.6.2 Simple Moving Averages
Disadvantage is that all values in the average are weighted
equally: the oldest value is given the same weight as the most
recent value.
34
Forecasting
1.6.3 Weighted Moving Averages
❑ It is similar to a simple moving average, except that it assigns
more weight to the most recent values in a time series.
That is, the highest weights are assigned to the most recent value
in the moving average, then the less high weight, to the next
most recent value, and so on. Note that the weights sum to 1.
35
Forecasting
1.6.3 Weighted Moving Averages
Ft = wn An + wn-1 A(n-1) + ………….. + w1 A1
Where, wn > wn-1 > …… > w1 are the n weights for the n points
in the average
Note that, ∑ wi = 1
36
Forecasting
Forecasting accuracy
The forecast error (also known as a residual) is the difference
between the actual value and the forecast value for the
corresponding period.
Et=A t- Ft
where Et is the forecast error at period t, At is the actual value at
period t, and Ft is the forecast for period t.
37
Forecasting
Example 1
Period Demand
Given the demand data for the last five 1 42
2 40
weeks, forecast demand for week 6 using 3 43
4 40
the following forecasting methods:
5 41
a) Three-points simple moving average,
b) Four-points weighted moving average using a weight of 0.4
for the most recent period, 0.3 for the next most recent, 0.2 for
the next, and 0.1 for the next.
c) If the actual for week 6 is 39, forecast demand for week 7
using the same weights as in part b). 38
Forecasting
Solution
a) The 3 most recent actual values are : 41, 40, and 43, then;
F6 = (41 + 40 + 43) / 3 = 41.33
If actual demand in week 6 turns to be 39, the moving average
forecast for week 7 would be:
F7 = (39 + 41 + 40) / 3 = 40
b) The 4 most recent actual values are: 41, 40, 43, and 40, then;
F6 = 41 x 0.4 + 40 x 0.3 + 43 x 0.2 + 40 x 0.1 = 41
c) F7 = 39x 0.4 + 41 x 0.3 + 40 x 0.2 + 43 x 0.1 = 40.2
39
Forecasting
1.6.4 Exponential Smoothing
Each new forecast is based on the previous forecast plus a
percentage of the difference between that forecast and the
actual value of the series at that point. That is:
New Forecast = Old Forecast + α (Actual - Old forecast)
Ft = Ft-1 + α (At-1 - Ft-1)
Or Ft = α. At-1 + (1 – α). Ft-1
Where; Ft = Forecast for period t
Ft-1 = Forecast for period t -1
α = Smoothing constant
At-1 = Actual demand or sales for period t -1
40
Forecasting
Example 2
The following table illustrates the actual demand for periods
from 1 to 11. Using the exponential smoothing, find the two
series of forecasts for each period. One forecast uses α = 0.1
and one uses α = 0.4. Naïve forecasts are used to develop the
forecasts for period t = 2.
41
Forecasting
α = 0.1 α = 0.4
Actual Forecast Error Forecast Error
Period t
Demand
Ft = Ft-1 Ft = Ft-1
At-1 - Ft-1 At-1 - Ft-1
+α.Error +α.Error
1 42 - - - -
2 40
3 43
4 40
5 41
6 39
7 46
8 44
9 45
10 38
11 40
12
42
Forecasting
α = 0.1 α = 0.4
Actual Forecast Error Forecast Error
Period t
Demand
Ft = Ft-1 Ft = Ft-1
At-1 - Ft-1 At-1 - Ft-1
+α.Error +α.Error
1 42 - - - -
2 40 42 -2 42 -2
3 43 41.8 1.2 41.2 1.8
4 40 41.92 - 1.92 41.92 - 1.92
5 41 41.73 - 0.73 41.15 - 0.15
6 39 41.66 - 2.66 41.09 - 2.09
7 46 41.39 4.61 40.25 5.75
8 44 41.85 2.15 42.55 1.45
9 45 42.07 2.93 43.13 1.87
10 38 42.35 - 4.35 43.88 - 5.88
11 40 41.92 - 1.92 41.53 - 1.53
12 41.73 40.92
43
Forecasting
Example 3
For the given sales data for a car each month, No. of
Month
cars
calculate forecast for the 11th month (F11), using:
1 40
❑ Naïve forecast
2 47
❑ Simple moving average MA . 3 43
❑ Four points moving average MA4 . 4 52
❑ Four points weighted moving average MAW4 5 59
6 64
using weights (0.4, 0.3, 0.2, 0.1)
7 62
❑ Exponential smoothing with error factor
8 65
α = 0.35 (smoothing factor). 9 68
10 72
44
Forecasting
Solution
a) Simple moving average :
F11 = MA =( 40 + 47 + 43 …..+72 ) / 10 = 58 cars .
b) Four points MA4:
F11 = MA4 = ( 72 + 68 + 65 + 62 ) / 4 = 66.75 = 67 cars.
c) Weighted MAw4 :
Ft = wn An + wn-1 A(n-1) + ………….. + w1 A1
F11 = (0.4) (72) + (0.3) (68) + (0.2) (65) + (0.1) (62)
= 68.4 = 69 cars
45
Forecasting
d) Exponential smoothing :
F11 = F10 + α (D10 – F10)
We use four points moving average, MA4 to
calculate F10
F10 = (68 + 65 + 62 + 64 ) / 4 = 64.75
F11 = 64.75 + 0.35 (72 – 64.75) = 67.29 = 68 cars
46
Forecasting
1.6.5 Trend forecasting
A linear trend equation is used to develop forecasts
when trend is present. It has the form:
yt = a + b t where;
t = Specified number of time periods from t = 0
yt = Forecast for period t
a = Value of yt at t = 0 - or the intercept with Y axi
b = Slope of the regression line
47
Forecasting
1.6.5 Trend forecasting
The equation can be plotted by finding two points on
the line. One can be found by substituting some
value of t into the equation and then solving for yt.
The other point is found by substituting another value
of t into the equation and then solving for the
corresponding yt
Plotting those two points and drawing a line through
them will yield a graph of the linear trend line.
48
Forecasting
1.6.5 Trend forecasting
The coefficients of the line, a, and b can be computed
from historical data using the following two
equations that are identical to those used for
computing a linear regression line:
b = (n ∑t y - ∑ t ∑ y) / (n ∑ t2 – (∑ t)2)
a = (∑ y - b ∑ t) / n
Where; n = Number of periods,
y = Value of the time series 49
Forecasting
week Unit sales
1.6.5 Trend forecasting
Example 4
1 700
Product sales of a firm over the last 10 weeks 2 724
3 720
are shown in the opposite table. Plot the data,
4 728
and visually check to see if a linear trend line 5 740
6 742
would Be appropriate. Then determine the 7 758
equation of the trend line, and predict sales 8 750
9 770
for weeks 11 and 12. 10 775
50
Forecasting
1.6.5 Trend forecasting Week
y ty t2
t
Solution
1 700 700 1
The following table shows 2 724 1448 4
The computation of different 3 720 2160 9
∑’s in the equations of the 4 728 2912 16
trend line coefficients, a, 5 740 3700 25
6 742 4452 36
and b
7 758 5306 49
8 750 6000 64
9 770 6930 81
10 775 7750 100
∑ = 55 7407 41358 51385
Forecasting
1.6.5 Trend forecasting
Solution
b = (n ∑t y - ∑ t ∑ y) / (n ∑ t2 – (∑ t)2)
= (10 x 41358 – 55 x 7407) / (10 x 385 – 552)
= 6159 / 825 = 7,51
a = (∑ y - b ∑ t) / n
= (7407 – 7.51 x 55) / 10 = 699.4
thus the trend line is yt = 699.4 + 7.51 t
52
Forecasting
1.6.5 Trend forecasting
Solution
by substituting values of t into this equation, we obtain
forecasts for future periods. For the next two periods
11 and 12, the forecasts are:
y11 = 699.4 + 7.51 x 11 = 782.01
y12 = 699.4 + 7.51 x 12 = 789.51
53
