Defining Baseline Sales in a Competitive Environment

Abstract

The baseline ("normal" sales) has become the most commonly used procedure

to measure the incremental effect of marketing variables, particularly

promotions. Firms such as IRI and Nielsen provide their clients with

baseline sales' estimates which they then use to determine the sales and

profit effects of consumer and trade promotions and, in some cases, measure

their brand equity. With the increasing importance of the baseline to the

measurement of marketing effects, it is necessary to develop an accurate

definition of the baseline and determine what it measures. This paper will

define baseline sales and discuss its implications, show that the incorporation

of competitive reaction is important to the definition of the baseline

and discuss how to use this concept when a firm has a broad product line. In

addition, we compare several baseline estimation methods using simulated

data and show the importance of incorporating a competitive reaction

component in baseline estimation. Finally, we apply the concept of competitive

reaction to actual sales data and show that the baseline computed without

a competitive reaction hnction is ~ i g ~ c a n tdliyff erent from the baseline

with a competitive reaction hnction.

This study investigates whether changes of oil price would have precedence,

exogeneity and causal prediction to the stock markets. The result

shows that changes of oil price have precedence over the stock returns in the

United States, Japan and Korea markets. And the evidence suggests that

there be exogeneity of oil to the stock markets because the stock returns can

be causally interpreted by the current or past changes of oil price during the

past two decades. Thus changes of oil price would contain any information

exploitable in forecasting the stock markets and have the predictive value of

leading indicators.