Journal of Econometrics 89 (1999) 293 — 315 Outlier robust analysis of long-run marketing effects for weekly scanning data Philip Hans Franses*, Teun Kloek, Andre´ Lucas Econometric Institute, Erasmus University Rotterdam, P.O. Box 1738, NL-3000 DR Rotterdam, Netherlands Department of Financial Sector Management, ECO/BFS, Vrije Universiteit, De Boelelaan 1105, 1081 HV Amsterdam, Netherlands Abstract We consider econometric modeling of weekly observed scanning data on a fast moving consumer good (FMCG), with a specific focus on the relationship between market share, distribution, advertising, price, and promotion. Such data can show non-stationary characteristics. Therefore, we use cointegration techniques to quantify the long-run effects of marketing efforts. Since weekly scanning data can contain aberrant observa- tions due to, e.g., out-of-stock situations or measurement errors, we favor an outlier robust cointegration method, which we outline in detail. In our illustrative FMCG example, we find different results across robust and non-robust methods for the long-run marketing effects. 1999 Elsevier Science S.A. All rights reserved. JEL classification: M31; C32; M32 Keywords: Market share; Distribution; Scanning data; Cointegration LM-test; Outlier robust method 1. Introduction An investigation of the long-run effects of marketing instruments, such as distribution, price, advertising, and promotion, is the subject of several recent * Corresponding author. E-mail: franses@few.eur.nl 0304-4076/99/$ — see front matter 1999 Elsevier Science S.A. All rights reserved. PII: S 0304-4076(98)00065-7