The Marketing Book 5th Edition

Quantitative methods in marketing 225

Table 9.10 Applications of statistical decision theory or stochastic

methods in marketing (approaches, advantages, limitations and when

recommended to use)

Method Based on Marketing applications Advantages Limitations

Queuing Probability

distribution analysis

of data (empirically

gathered on how the

major factors/

variables will affect

the situation-

problem under

analysis). It is an

analysis of queuing

systems attempting

to determine service

levels/performance

(a) Optimize:

salesforce (Paul,

1972), number

of checkouts,

number of

attendants, etc.

(b) Minimize

inventory-

carrying costs;

suitable and

widely used by

chain stores,

supermarkets,

department

stores, petrol

stations, airline

ticket offices,

ports, airports,

etc.

(a) Predicts how

different

marketing

systems will

operate

(b) Gives explicit

expression

relating the

design of a

system to the

length and

frequency of

queues, waiting

time, etc.

(a) Must be

operated a

sufficient length

of time to

achieve a steady-

state solution

(b) Manager’s

reluctance to

have confidence

in this method

Stochastic

process

A random

experiment which

occurs over time

and whose outcome

is determined by

chance. This is an

analysis of systems

with variable/

uncertain

components

(a) For building

choice, models

checking on

customers’

loyalty

(b) Predict buying

decisions and

future purchasing

probabilities

Might predict flow of

customers and

future purchase

probabilities

Suitable for short-

run predictions only

Statistical

decision

theory

This is an analysis of

decision-making

processes where

outcomes are

uncertain. Probability

of each outcome –

based upon past

data or subjective

estimates – is given

adequate weight and

is taken into

consideration for

decision-making

For decision-making

on: budgeting,

advertising, pricing,

test-marketing, new

product

development,

merchandising,

optimum mix, etc.

Simplifies the level of

analysis and suggests

a number of possible

outcomes

Subjective estimation

of the probability for

each decision might

affect the results’

validity