Unit 2 Forecasting
Why forecasting
• The decision-makers of an organisation need
to allocate limited resources(time, money,
marketing resources, etc.) to different areas of
the business, and this needs to be done in a
logical and efficient way.
• In a highly complex and dynamic world,
decision-makers need to find a wayto deal
with the complexity and uncertainty that is
associated with such an environment.
1. Gather 2. Compile a
information forecast
4. Incorporate new 3. Monitor forecast
information vs. actual
Gathering information
Quantitative information • Qualitative information
is simply information that is simply information
can be measured that has no direct
mathematically in terms financial consequence
of absolute value or but may affect the
change. accuracy of the forecast
or the assumptions
upon which the forecast
is built.
Quantitative
• value (the amount of money that is expected
to be spent or received);
• timing (when and for how long the money will
be spent or received);
• volatility (how rapidly and regularly does the
expected amount change); and
• relevance (is the expected amount related to
the decision at hand).
Example 2.1 VALUE EMPHASIS
• 55000 boxes X 16 X 0.25
• 220 00 kg =$ 12 540 000
• $ 12 540 000 x 1.05 = $13 167 000 if produce same
amount
• Per Kg = $12 540 000/220 000 = $ 57 KG
• Up to $57x 1.05 = $ 59.85
• NEW MARKET SHARE AT 75 000 BOXES
• Now 75 000x $ 59.85 = $ 17 995 000
• COMBINATION OF INFLATION IN PRICE AND VOLUME
IN GROWTH
EXAMPLE 2.1
• Timing also a factor to be taken into
consideration
• Volatility : measure of volatility
• Relevance
Qualitative
• no direct financial consequence, but may
affect the accuracy of the forecast or the
assumptions upon which the forecast is built
APPROACHES TO CONSIDER
• TREND ANALYSIS
– Example 2.3 page 77-78
• SWOT ANALYSIS
– SW internal/current
– OT external/future
• Cost benefit
– Accuracy vs. cost
RELATIONSHIPS BETWEEN VARIABLES
• important to • scatter diagrams
understand the • linear regression model
relationship between • correlation coefficient
two (or more) variables
• E.g. Marketing budget
and sales relationship
• QUESTION : What is the
strength of the
relationship
Scatter diagram
• May show some kind of relationship and trend
• Does not show causal relationship; change in
one causes a change in the other with a
discernable pattern
LINEAR REGRESSION
Variables
Dependent variables are uncontrollable in
nature, and are directly influenced by changes in
independent variables. Dependent variables can
includes metrics such as sales revenue, sales
growth, new-customer inquiries, same-store
repeat sales, and the average size of
transactions. The outcomes of dependent
variables reveal whether certain strategies are
more or less effective than others, allowing
companies to invest in the most effective ideas.
Measure a set of chosen dependent variables
that most directly influence your strategic
marketing goals.
Independent variables are the controllable factors
that marketing researchers use to influence
changes in the values of dependent variables. The
elements of the marketing mix -- price, product,
place and promotion -- are independent variables,
since marketers can intentionally alter these factors
at any time. The purpose of a marketing mix
strategy, and thus of marketing research, is to
maximize the outcome of dependent variables by
finding the perfect combination of values for
independent variables. Alter as many independent
variables as you can to discover impacts and
interrelationships that you might not have
expected.
SPREADSHEET EXAMPLE OF LINEAR
REGRESSSION
LINEAR REGRESSION ASSUMPTIONS
• There is a linear relationship between the
dependent and independent variable.
• There is only one independent variable that drives
the dependent variable
• Past trends in the relationship between the
dependent and independent variable will continue
into the future
• The available data are sufficient to accurately
calculate the relationship between two variables
CORRELATION COEFFICIENT
