lecture 6 Response models and elasticities Flashcards
Relevance of sales response models
looking for the relationship between advertising and sales is somewhat worse than looking for a needle in a haystack
Business managers need to know how markets respond to the actions they take
One of the primary goalds of marketing science is to provide a structural insight of how a brand generates its sales, market share, customer awareness or any other variable of interest
Sales response model
Model: is a generally simplified mathematical representation of real world relations
Sales response: is the rate at which sales change as a result of changes in business activity
Sales response models: try to model a sales response as a function of business activities
markets data and variables
before modeling a response function, the level of analysis should reflect the heterogeneous nature of a market
The firm conducts certain business activities and communicates with consumers through market activities and price settings. The customers respont to actions and inform company about needs, and based off those needs the company provides products and services. As a reaction customers produces sales for the firm as a reaction.
Internal data
data comes from company itself
Advantage: usually readily available, intra-daily basis, accurate and quick
Disadvantages:
Only from company itself - no competitors, no information about psychographics, no information about time of decision
Household scanner data
data coming from retail upc scanners or from company itself
Advantages: available for all major retail chains, available on a daily basis
Disadvantages:
Sampling frame (only “shops”), small shops may not be considered, no information about psychographics, no information about time of decision
Annual reports (e.g. compustat)
data coming from annual reports of companies
Advantages: available for all companies that are listed on a stock market
Disadvantage:
Data is only available on a yearly (or quarterly) basis, non-traded companies are not considered, no information about psycholgraphics, no information about time of decision
Sales response model
capture the factors that drive the market
sales = f(1) (quality, price, advertising…(2)
1) what are the relevant sales drivers
2) what functional form appropriately represents the manner in which the drivers exert their effect
overview of the types of static sales response models
1) connstant marginal returns (linear)
2) decreasing marginal returns (multiplicative model, semi-logarithmic model)
3) Saturation volume (modified exponential model)
4) S-shaped (log-reciprocal model, logistic model
5) market share model (multiplicative interaction model (MCI), multinominal logit model (MNL))
constant marginal returns
Linear regression
Objective of regression analysis:
quantification of the slope of a regression line
Estimation of the influence of one variable (X) on another variable (Y)
R^2 expresses the proportion of the explained variance in the dependent variable (Y) that is explained by the regression ine (value range: 0 to 1)
Linear models assume constant returns to scale
This means that each additional unit in some X (advertising) will lead to an equal incremental change in Y (Sales)
Elasticity
e = dQ/dp * p/Q = deltaQ(%)/deltap(%)
The price elasticity of demand tells us how sensitive peoples purchases are of a product to changes in the price of the product
income elasticity of demand
how sensitive peoples purchases of a product are in response to changes in their incomes
if your income rises by 10% you consume 15% more music, 10% more pizza and 5% less ramen noodles what does this make these products
music = superior
pizza = normal
Ramen noodles = inferior
absolute effect
change in sales / change in advertising euro
==
sales now - sales before / detailing eur now - detailing eur before
Relative effect (elasticity)
change in sales %/ advertising eur in %
==
(salesnow - salesbefore/sales before )/Detailing eur now - detailing eur before /detailing eur before)
Multiplicative model
Q = a0 *x1^a1 *eû
Ex = a1
