Demand Forecasting

Qualitative Methods

10 Sep
2009
Abhijith R Kashyap
07EC01

Engineering Economics Assignment 1
Qualitative Methods of Demand Forecasting

Introduction

Supply and demand is an economic model based on price, utility and quantity in a market. It concludes
that in a competitive market, price will function to equalize the quantity demanded by consumers, and
the quantity supplied by producers, resulting in an economic equilibrium of price and quantity.
In order to achieve this equilibrium, the producer has to have optimum output, which matches the
quantity demanded by the consumer, as closely as possible. Thus forecasting the demand is an essential
component of all manufacturing activities.
Demand Forecasting is the activity of estimating the quantity of a product or service that consumers will
purchase. Demand forecasting involves techniques including both informal methods, such as educated
guesses, and quantitative methods, such as the use of historical sales data or current data from test
markets. Demand forecasting may be used in making pricing decisions, in assessing future capacity
requirements, or in making decisions on whether to enter a new market.
Thus the types of Demand forecasting may be classified as
1. Qualitative Forecasting
2. Quantitative Forecasting




Qualitative demand forecasting is based on expert opinion. It assumes certain aspects of market behavior
and simulates the sale of the commodity under trial. Some of the types in this method are,
Unaided judgment
Prediction market
Delphi technique
Game theory
Judgmental bootstrapping
Simulated interaction
Intentions and expectations surveys
Conjoint analysis

Qualitative demand forecasting is based on statistical data collected from the market and extrapolating
the information suitably.
Methods that rely on quantitative data
Discrete Event Simulation
Extrapolation
Quantitative analogies
Rule-based forecasting
Neural networks
Data mining
Causal models
Segmentation


Need for Demand Forecasting

For a producer, there exist two approaches towards ascertaining the quantity that is to be produced.
1. Forecasting Sales
2. Demand Forecasting
Sales forecast form the basis for retailers to keep the required stock. They form the basis for buying
decisions and dictate operational planning, such as replenishment and labor scheduling. Sales forecasts
are also essential to financial planning and to all decision-making that flows from financial plans.
However, if you rely only on sales forecasts, you may fail to exploit the full value of demand for your
merchandise. You will be unprepared for the level of consumer demand, and youll have insufficient
inventory levels, resulting in lost sales. Or, you will have excess levels of inventory, significantly raising
your cost per item and decreasing revenue.

Having a clear idea of the demand for the product, adequate measures can be taken to publicize the
product. If a new product has extremely high demand, much more than the maximum supply, then the
advertisements are to be kept at minimum to avoid consumer dissatisfaction.
A similar argument can be made for commodities with lower expected demand. Publicity for these
products would help in achieving the market equilibrium much faster and without losses of production.



Factors affecting Demand

Stock effects

The effects that inventory levels have on sales. In the extreme case of stock-outs, demand coming into
your store is not converted to sales due to a lack of availability. Demand is also untapped when sales for
an item are decreased due to a poor display location, or because the desired sizes are no longer
available. For example, when a consumer electronics retailer does not display a particular flat-screen TV,
sales for that model are typically lower than the sales for models on display. And in fashion retailing,
once the stock level of a particular sweater falls to the point where standard sizes are no longer
available, sales of that item are diminished.


Market and
Category Forecast
Actions Of
Insiders
Actions of
Insiders
Market Share
Demand
Market response effects

The effect of market events that are within and beyond a retailers control. Demand for an item will
likely rise if a competitor increases the price or if you promote the item in your weekly circular. The
resulting sales increase reflects a change in demand as a result of consumers responding to stimuli that
potentially drive additional sales. Regardless of the stimuli, these forces need to be factored into
planning and managed within the demand forecast.

In this case demand forecasting uses techniques in causal modeling. Demand forecast modeling
considers the size of the market and the dynamics of market share versus competitors and its effect on
firm demand over a period of time. In the manufacturer to retailer model, promotional events are an
important causal factor in influencing demand. These promotions can be modeled with intervention
models or use a consensus process to aggregate intelligence using internal collaboration with the Sales
and Marketing functions.

Unlike sales, demand represents the relationship between item and quantity purchase intent, and the
many variables that influence a consumers purchased decisions. These variables include:
Item price
Related item prices offered by the retailer or its competitor
Promotional tactics
In store merchandising tactics
Time of day, week or year.
To manage demand effectively and efficiently, it is important to understand, anticipate and fully uncover
the effects of these market forces on potential sales. This level of analysis and preparation will allow one
to address the effects of drivers outside your control and based on insight and fully support your
business objectives.
Forecasting demand allows you to manage and prepare for the many forces on item, assortment and
category demand.


Qualitative Demand forecasting

1. Unaided Judgment

It is a common practice to ask experts what will happen. This is a good procedure to use when
Experts are unbiased
Large channels are unlikely
Relationships are well understood
Experts possess privileged information
Experts receive accurate feedback about their forecasts
Unfortunately, unaided judgment is often used even when many of the above guidelines do not
hold good. As a result the accuracy of the forecast is affected.
2. Prediction Markets
Prediction markets are also known as betting markets, and future markets have a long history.
Between the end of the Civil war of the USA and World War II, well organized markets for
betting on presidential elections correctly picked the winner of every case but 1916.
Some commercial organizations provide internal markets and software that allows participants
to bet by trading contracts. Confidential betting markets can also be setup within firms to bet on
such things as the sales growth of a new product. However there are no empirical studies that
compare forecasts from prediction markets with those from traditional groups or from other
methods.
3. Delphi Technique

To forecast with Delphi, the administrator then provides the experts with anonymous summary
statistics on the forecasts and experts reasons for their forecasts. The process is repeated until
there is little change in the forecasts between the rounds. The Delphi forecast is the median or
mode of the experts final forecasts.

4. Structural Analogies

The outcomes of similar situations from the past may help a marketer to forecast the outcome
of a new situation. For example, the introduction of new products in the European market can
provide analogies for outcome of subsequent releases in other countries. To use the structural
analogies method, the administrator prepares a description of the target situation and selects
experts who have knowledge of analogous situations, preferably direct experience. The experts
identify and describe analogous situations, rate their similarity to the target situation and match
the outcomes of their analogies with potential outcomes in the target situations. The
administrator then devices forecasts from the information the experts provided on their most
similar analogies.

5. Game theory

Game theory is touted as a way to obtain better forecasts in situations involving negotiations
and other conflicts. Game theory is a general mathematical theory which tries to single out the
best possible outcome from an event. Every event is treated as a game, in which the players
may choose their path to the finish. The general idea is that, maximizing the outcome for each
individual player may not yield the best outcome for the game as a whole. So the broader
picture of the game, in this case the market is to be kept in consideration.
For example, say a distributor has to give the retailer the stock. The best way is to provide the
populous retailer more stock rather than giving everyone equal stock.
Since this is a highly mathematical approach, the real world factors are very hard to include and
hence certainty of forecasts will not be very high.

6. Judgmental Decomposition
The basic idea behind judgmental decomposition is to divide the forecasting problem into parts
that are easier to forecast than the whole. One then forecasts the part individually. Finally the
parts are combined to obtain a forecast. For example to forecast the sales for a brand, one can
forecast industry sales volume, market share, and selling price per unit. Then reassemble the
problem by multiplying the components together. Empirical results indicate that, in general,
forecasts decompositions are more accurate where there is much uncertainty about the
aggregate forecasts and where large numbers are involved.

7. Judgmental Bootstrapping
Judgmental bootstrapping is a type of expert system. It translates an expert's rules into a
quantitative model by regressing the experts forecasts against the information that he used.
Bootstrapping models apply an expert's rules consistently, and many studies have shown that
decisions and predictions from bootstrapping models are similar to those from the experts.
Three studies showed that bootstrapping improved the quality of production decisions in
companies. To date, research on forecasting with judgmental bootstrapping has been restricted
primarily to cross-sectional data, not time-series data. Studies from psychology, education,
personnel, marketing, and finance, showed that bootstrapping forecasts were more accurate
than forecasts made by experts using unaided judgment. They were more accurate for eight of
eleven comparisons, less accurate in one, and there were two ties. The gains in accuracy were
generally substantial. Bootstrapping can be useful when historical data on the variable to be
forecast are lacking or of poor quality; otherwise, econometric models should be used.
Bootstrapping is most appropriate for complex situations, where judgments are unreliable, and
where experts judgments have some validity. When many forecasts are needed, bootstrapping
is cost-effective. If experts differ greatly in expertise, bootstrapping can allow one to draw upon
the forecasts made by the best experts. Bootstrapping aids learning; it can help to identify
biases in the way experts make predictions, and it can reveal how the best experts make
predictions. Finally, judgmental bootstrapping offers the possibility of conducting experiments
when the historical data for causal variables have not varied over time. Thus, it can serve as a
supplement for econometric models.
8. Expert Systems

Expert systems forecasting involves identifying forecast rules set by the experts and rules
learned from empirical research. One should aim for simplicity and completeness in the
resulting system and the system should explain the forecasts to the users. Developing an expert
system is expensive and so the method will only be of interest in situations where many
forecasts of similar kind are required.


9. Simulated Interactions
Simulated interaction is a form of role playing for predicting decisions by people who are
interacting with others. It is especially useful when the situation involves conflict. For example,
one might wish to forecast how best to secure an exclusive distribution arrangement with a
major supplier. To use simulated interaction, an administrator prepares a description of the
target situation, describes the main protagonists roles, and provides a list of possible decisions.
Role players adopt a role and read about the situation. They then improvise realistic interactions
with the other role players until they reach a decision; for example to sign a trial one-year
exclusive distribution agreement. The role players decisions are used to make the forecast.
Using eight conflict situations, Green (2005) found that forecasts from simulated interactions
were substantially more accurate than can be obtained from unaided judgment. Simulated
interaction can also help to maintain secrecy.

10. Intentions and expectations surveys
With intentions surveys, people are asked how they intend to behave in specified situations. In
a similar manner, an expectations survey asks people how they expect to behave. Expectations
differ from intentions because people realize that unintended things happen. For example, if
you were asked whether you intended to visit the dentist in the next six months you might say
no. However, you realize that a problem might arise that would necessitate such a visit, so your
expectations would be that the event had a probability greater than zero. This distinction was
proposed and tested by Juster (1966) and its evidence on its importance was summarized by
Morwitz (2001). Expectations and intentions can be obtained using probability scales such as
Justers eleven-point scale. The scale should have descriptions such as 0 = No chance, or almost
no chance (1 in 100) to 10 = Certain, or practically certain (99 in 100).
To forecast demand using a survey of potential consumers, the administrator should prepare an
accurate and comprehensive description of the product and conditions of sale. He should select
a representative sample of the population of interest and develop questions to elicit
expectations from respondents. Bias in responses should be assessed if possible and the data
adjusted accordingly. The behavior of the population is forecast by aggregating the survey
responses.
Useful methods have been developed for selecting samples, obtaining high response rates,
compensating for non-response bias, and reducing response error. Dillman (2000) provides
advice for designing surveys. Response error (where respondent information is not accurately
reported) is probably the largest component of total error for marketing
problems.

By surveying consumers about their preferences for alternative product designs in a structured
way, it is possible to infer how different features will influence demand. Potential customers
might be presented with a series of perhaps 20 pairs of offerings. For example, various features
of a personal digital assistant such as price, weight, battery life, screen clarity and memory could
be varied substantially such that the features do not correlate with one another. The potential
customer is thus forced to make trade-offs among various features by choosing one of each pair
of offerings in a way that is representative of how they would choose in the marketplace. The
resulting data can be analyzed by regressing respondents choices against the product features.
The method, which is similar to bootstrapping, is called conjoint analysis because respondents
consider the product features jointly.

In general, the accuracy of forecasts from conjoint analysis is likely to increase with increasing
realism of the choices presented to respondents (Wittink and Bergesteun 2001). The method is
based on sound principles, such as using experimental design and soliciting independent
intentions from a sample of potential customers. Unfortunately however, there do not appear
to be studies that compare conjoint-analysis forecasts with forecasts from other reasonable
methods.



1. Conjoint Analysis
Conclusions

Significant gains have been made in forecasting for marketing, especially since 1960. Advances have
occurred in the development of methods based on judgment, such as Delphi, simulated interactions,
intentions studies, opinions surveys, bootstrapping, and combining. They have also occurred for
methods based on statistical data, such as extrapolation, rule-based forecasting, and econometric
methods. Most recently, gains have come from the integration of statistical and judgmental forecasts.
General principles
Managers domain knowledge should be incorporated into forecasting methods.
When making forecasts in highly uncertain situations, be conservative. For example, the trend
should be dampened over the forecast horizon.
Complex methods have not proven to be more accurate than relatively simple methods. Given
their added cost and the reduced understanding among users, highly complex procedures
cannot be justified.
When possible, forecasting methods should use data on actual behavior, rather than judgments
or intentions, to predict behavior.
Methods that integrate judgmental and statistical data and procedures (e.g., rule-based
forecasting) can improve forecast accuracy in many situations.
Overconfidence occurs with quantitative and judgmental methods.
When making forecasts in situations with high uncertainty, use more than one method and
combine the forecasts, generally using simple averages.

Methods based on judgment
When using judgment, rely on structured procedures such as Delphi, simulated interaction,
structured analogies, and conjoint analysis.
Simulated interaction is useful to predict the decisions in conflict situations, such as in
negotiations.
In addition to seeking good feedback, forecasters should explicitly list all the things that might
be wrong about their forecast. This will produce better calibrated prediction intervals.

Finally, efforts should be made to ensure forecasts are free of political considerations in a firm. To help
with this, emphasis should be on gaining agreement about the forecasting methods. Also, for important
forecasts, decisions on their use should be made before the forecasts are provided. Scenarios are helpful
in guiding this process.
References
Demand Forecasting : Wikipedia
o http://en.wikipedia.org/wiki/Demand_Forecasting
Demand Forecasting: Evidence-based Methods
o J. Scott Armstrong and Kesten C. Green
Forecasting Demand: Monitoring Demand Leads to More Profitable Decision-
Making
o Oracle white paper