FORECASTING

What is FORECASTING

Forecast is a statement about the future. It also serves as a basis

of planning.
Two uses for forecasts:
1. To help managers plan the system. (long-range plans)
2. To help them plan the use of the system. (short-range and
intermediate-range plans)
 Features Common to All Forecasts

 Forecasting techniques generally assume that the same underlying causal

system that existed in the past will continue to exist in the future.
 Forecasts are rarely perfect; actual results usually differ from predicted
values.
 Forecasts for groups of items tend to be more accurate than forecasts for
individual items because forecasting errors among items in a group
usually have a canceling effect.
 Forecasts accuracy decreases as the time period covered by the forecast
– the time horizon increases.
 Basic Forecasting Approaches

QUANTITATIVE QUALITATIVE
Historical data from Opinions from experts
time-series or decision makers, or
correlation information customers
 DEFINITION of Terms

Stationary time series – A time series whose statistical properties are independent of
time. For a stationary time series, the process generating the data has a constant mean
and the variability of the time series is constant over time.
Trend pattern – A trend pattern exists if the time series plot shows gradual shifts or
movements to relatively higher or lower values over a longer period of time.
Seasonal pattern – A seasonal pattern exists if the time series plot exhibits a repeating
pattern over successive periods.
Cyclical pattern – A cyclical pattern exists if the time series plot shows an alternating
sequence of points below and above the trend line lasting more than one year.a
Forecast error – The difference between the actual time series value and forecast.
 DEFINITION of Terms

Mean absolute error (MAE) – The average of the absolute values of the forecast
errors.
Mean squared error (MSE) – The average of the sum of squared forecast errors.
Mean absolute percentage error (MAPE) – The average of the absolute values of
the percentage forecast errors.
Moving averages – A forecasting method that uses the average of the k most recent
data values in the time series as the forecast for the next period.
Weighted moving averages – A forecasting method that involves selecting a
different weight for the k most recent data values in the time series and then
computing a weighted average of the values. The sum of the weights must equal one.
 DEFINITION of Terms

Exponential smoothing – A forecasting method that uses a

weighted average pf past time series values as the forecast; it is a
special case of the weighted moving averages method in which
we select only one weight – the weight for the most recent
observation.
Smoothing constant – A parameter of the exponential smoothing
model that provides the weight given to the most recent time
series value in the calculation of the forecast value.
Regression analysis – A procedure for estimating values of a
dependent variable given the values of one or more independent
variables in a manner that minimizes the sum of the squared errors.
 DEFINITION of Terms

Dependent variable – The variable that is being predicted or

explained in a regression analysis.
Independent variable – A variable used to predict or explain
values of the independent variable in regression analysis.
Categorical (dummy) variable – A variable used to categorize
observations of data. Used when modeling a time series with a
seasonal pattern.
 Quantitative Forecasting Methods

Quantitative
Methods

Associative Models
Time-Series Models
Associative models (often called causal
Time series models look at past
models) assume that the variable being
patterns of data and attempt to predict
forecasted is related to other variables in
the future based upon the underlying
the environment. They try to project
patterns contained within those data.
based upon those associations.
 Forecasts Based on Judgment and Opinion

Judgmental Forecasts are forecasts that use subjective inputs such as opinions from
consumer surveys, sales staff, managers, executives, and experts. Forecasts are based
on:
1. Executive Opinion
2. Delphi Method
3. Sales Force Estimates
4. Consumer Surveys
5. Outside Opinion
6. Opinions of Managers and Staff
Delphi Method. Managers and staff complete a series of questionnaires, each
developed from the previous one, to achieve a consensus forecast.

Reasons:
1. The group of experts can provide needed judgmental input.
2. More individuals may be needed that can interact effectively in a face-to-face
situation, and/or the individuals cannot be conveniently assembled in one place.
3. It is important to avoid a “bandwagon effect”
4. It is desirable to preserve the anonymity of the participants.
 Forecasts Based on Time Series Data

Time series. A time-ordered sequence of observations taken at regular intervals over time.
Forecasting techniques based on time series data are made on the assumption that future values of the series
can be estimated from the past values.
Data may be measurements of demand, earnings, profits, shipments, accidents, output, precipitation, productivity
and the consumer price index.
Analysis of time series data requires the analyst to identify the underlying behavior of the series. These
behaviors can be described as follows:
1. Trend – a long-term movement in data.
2. Seasonality – short-term regular variations related to weather or other factors.
3. Cycle – wavelike variation lasting more than one year.
4. Irregular variations – caused by unusual circumstances, not reflective of typical behavior.
5. Random variations – residual variations after all other behaviors are accounted for.
Techniques of
Averaging
1. Naive Approach – the forecast for any period equals the previous period’s actual value.
Ft+1 = Xt
Where
F = forecast
t = time period
X = observed value

Advantages:
1. It has virtually no cost.
2. It is quick and easy to prepare.
3. It is easily understandable.
Disadvantages:
1. The forecast just traces the actual data, with a lag of one period.
2. It does not smooth at all.
2. Moving Average – technique that averages a number of recent actual values, updated as new values become
available.
n
ΣAi
Man = i=1
n
where
i = age of the data
n = number of periods in the moving average
A = actual value with age i
MA = forecast

Advantages:
1. It is easy to compute.
2. It is easy to understand.
Disadvantages:
1. All values in the average are weighted equally.
Weighted Average – an average in which each quantity to be averaged is assigned a weight.
3. Exponential Smoothing - weighted averaging method based on previous forecast plus a percentage of
the forecast error.
Ft = Ft-1 + (At-1 – Ft-1)
where
Ft = forecast for period t
Ft-1 = forecast for period t-1
= smoothing constant
At-1 = actual demand or sales for period t-1
Smoothing constant  represents a percentage of the forecast error.

The sensitivity of forecast adjustment to error is determined by the smoothing constant, . The closer its
value is to zero, the slower the forecast will be to adjust to forecast errors. The closer the value of  is
to 1.00, the greater the sensitivity and the less the smoothing.
Advantages:
Its minimal data storage requirements and ease of calculation.
The ease with which the weighting scheme can be altered.
Techniques of Trend
Linear Trend Equation – used to develop forecasts when trend is present.
yt = a + bt
where
t = specified number of time periods from t = 0
yt = forecast for period t
a = value of yt at t = 0
b = slope of the line

b = nΣty – Σt Σy
nΣt2 – (Σt)2
a = Σy – bΣt
n
where
n = number of periods
y = value of the time series
2. Trend-Adjusted Exponential Smoothing – variation of exponential
smoothing used when a time series exhibits trend.

AFt+1 = St + Tt
St = TAFt + 1 (At – TAFt)
Tt = Tt-1 + 2 (TAFt– TAFt-1 – Tt-1)

where
St = smoothed forecast
Tt = current trend estimate
 Techniques for Seasonability

Seasonal variations. Regularly repeating movements in series values that can be tied to
recurring events.
Seasonality in a time series is expressed in terms of the amount that actual values deviate
from the average value of a series. There are two different models of seasonality, additive
and multiplicative. In the additive model, seasonality is expressed as a quantity, which is
added or subtracted from the series average in order to incorporate seasonality. In the
multiplicative model, seasonality is expressed by the value of series to incorporate
seasonality. The seasonal percentages in the multiplicative model are referred to as
seasonal relatives or seasonal indexes.
Seasonal relatives. Seasonal relatives are used in two different ways in forecasting. One
way is to deseasonalize data; the other way is to incorporate seasonality in a forecast.
Deseasonalizing data can be accomplished by dividing each data point
by its corresponding seasonal relative. Incorporating seasonality can
accomplished in this way: (1) obtain trend estimates for desired periods
using a trend equation; (2) add seasonality to the trend estimates by
multiplying these trend estimates by the corresponding seasonal relative.

Computing Seasonal Relatives. A commonly used method for

representing the trend portion of a time series involves a centered moving
average. Centered moving average is a moving average positioned at
the center of the data that were used to compute it.
 Techniques for Cycles

The most commonly used approach is explanatory: search for another variable that relates to,
and leads, the variable of interest.
Techniques for Time Series
Box-Jenkins Technique. Increase popularity and ability to provide accurate forecasts.
Advantage is better able to handle data that include complex patterns than the techniques.
Disadvantages are its processing costs and complexity.
Predictor variables. Variables that can be used to predict values of the variable of interest.
Regression is a technique for fitting a line to a set of points.
Simple Linear Regression – the simplest and most widely used form of regression involves a
linear relationship between two variables.
Least Square Line – minimizes the sum of the squared deviations around the line.
yc = a + bx
where
yc = predicted (dependent) variable
x = predictor (independent) variable
b = slope of the line
a = value of yc when x = 0

The coefficients a and b of the line are computed

b = n(xy) – (x) (y)
n(x²) – (x)²

a = y – bx or y – bx
n
where
n = number of paired observations
Correlation measures the strength and direction of relationship between
two variables.
A correlation of +1.00 indicates that changes in one variable are always
matched by changes in the other.
A correction of -1.00 indicates that increases in one variable are
matched by decreases in the other.
A correlation close to zero indicates little linear relationship between
two variables.
r= n(xy) – (x)( y)
___________________________
n(x²) – (x)² • n(y²) – (y)²
Comments on the Use of Linear Regression Analysis
Use of simple regression analysis implies that certain assumptions have been satisfied.
1. Variations around the line are random.
2. Deviations around the line should be normally distributed.
3. Predictions are being made only within the range of observed values.
If the assumptions are satisfies, regression analysis can be a powerful tool. To obtain the best results, observe the following:
1. Always plot the data to verify that a linear relationship is appropriate.
2. The data may be time-dependent.
3. A small correlation may imply that other variables are important.
These weaknesses of regression:
1. Simple linear regression applies only to linear relationships with one independent variable.
2. A considerable amount of data is needed to establish the relationship – in practice, 20 or more variables.
3. All observations are weighted equally
 Accuracy and Control of Forecasts

Forecast error is the difference between the actual value and the value that was
predicted for a given period.
Positive errors result when the forecast is too low, negative errors when the
forecast is too high. Forecasts errors influence decisions in two somewhat different
ways. One is in making a choice between various forecasting alternatives, and the
other is in evaluating the success or failure of a technique in use.
Summarizing Forecast Accuracy
Two aspects of forecast accuracy can have potential significance when deciding
among forecasting alternatives. One is historical error performance of a forecast,
and the other is the ability of a forecast to respond to changes.
Two commonly used measures for summarizing historical errors MAD and MSE.
Mean absolute deviation. The average absolute error.
MAD =  (Actual – forecast) / n
Mean squared error. The average of squared errors.
MSE =  (Actual – forecast)2 / n - 1
Controlling the Forecast
There are a variety of possible sources of forecast errors, including the following:
The model may be inadequate due to (a) omission of an important variable, (b) a change or shift in
the variable that the model cannot deal with or (c) appearance of a new variable.
Irregular variations may occur due to severe weather or other natural phenomena, temporary
shortages or breakdowns, catastrophes, or similar events.
The forecasting technique may be used incorrectly or the results misinterpreted.
There are always random variations in the data.

Forecast can be monitored using either tracking signals or control chart.

Tracking signal. The ratio of cumulative forecast error to the corresponding value of MAD, used to monitor a
forecast.
Tracking signal = (actual – Forecast) / MAD

Bias. Persistent tendency for forecasts to be greater or less than the actual values of a time series.
MADt = MADt-1 + ([Actual – Forecast]t - MADt-1)

Control chart. Monitoring approach that sets limits for individual forecast errors; the limits are multiples of the
square root of MSE.
s = MSE

Choosing a Forecasting Technique

The manager or analyst must take a number of factors into consideration. The two most important factors are cost
and accuracy. Other factors to consider in selecting a forecasting technique include the availability of historical data;
the availability of computers; the ability of decision makers to utilize certain techniques; the time needed to gather
and analyze data, and prepare the forecast; and any prior experience with a technique.
Elements of a Good Forecast
The forecast should be timely.
The forecast should be accurate and the degree of accuracy should be
stated.
The forecast should be reliable, it should work consistently.
The forecast should be expressed in meaning units.
The forecast should be in writing. The forecasting technique should be simple
to understand and use.

Using Forecast Information

A manager can take a reactive or a proactive approach to a forecast.
A reactive approach views forecasts as probable descriptions of future
demand, and a manager reacts to meet that demand. A proactive approach
seeks to actively influence demand.
OPERATIONS STRATEGY

Forecasts are the basis for planning. Clearly, the more accurate an organization’s forecasts, the better
prepared it will be to take advantage of future opportunities and to reduce potential risks. Maintaining
accurate, up-to-date information on prices, demand, and other variables can have a significant impact on
forecast accuracy.

An organization also can do other things to improve forecasts. These do not involve searching for
improved techniques, but relate to the inverse relation of accuracy to the forecast horizon: forecasts that
cover shorter time frames tend to be more accurate than longer-term forecasts. Recognizing this,
management might choose to devote efforts to shortening the time horizon that forecasts must cover.
Essentially, this means shortening the lead time needed to respond to a forecast. This might involve
building flexibility into operations to permit rapid response to changing demands for products and
services, or to changing volumes in quantities demanded: shortening the lead time required to obtain
suppliers, equipment, and raw materials or the time needed to train or retrain employees; or shortening
the time needed to develop new products and services.
Thank you!