The attempt to estimate the effectiveness of advertising by means of standard econometric models often fails because (i) data are contaminated, (ii) the preconditions for the applicability of the statistical model are not met, (iii) the underlying economic model is defective. These sources of difficulties are discussed in detail, some suggested ways of avoiding the difficulties are examined, and the future of such approaches is assessed.
Get full access to this article
View all access options for this article.
References
1.
BoxG. E. P., and HunterJ. S., “A Confidence Region for the Solution of a Set of Simultaneous Equations with an Application to Experimental Design,”Biometrika, 41(1954), 190–199.
2.
FloodW., and KoopmansT. C., Studies in Econometric Method, Cowles Foundation Monograph No. 14, New York: John Wiley & Sons, 1953.
3.
JohnsonJ., Econometric Methods, New York: McGraw-Hill, 1963.
4.
MaffeiR. B., “Brand Preferences and Simple Markov Processes,”Operations Research, VIII (1960), 210–218.
5.
QuandtR. E., “Probabilistic Errors in the Leontief System,”Naval Research Logistics Quarterly, 5(1958), 155–170.
6.
StoneR., The Measurement of Consumers’ Expenditure and Behavior in the United Kingdom 1920–1938, Cambridge: Cambridge University Press, 1954.
7.
TelserL. G., “The Demand for Branded Goods as Estimated from Consumer Panel Data,”Review of Economics and Statistics, XLIV (1962), 300–324.
8.
TintnerG., Econometrics, New York: John Wiley & Sons, 1952.
9.
WinterPeter R., “Forecasting Sales by Exponentially Weighted Moving Averages,”Management Science, VI (1960), 324–42.
10.
WoldH., and JureenL., Demand Analysis, New York: John Wiley & Sons, 1953.
