Gaver, Kenneth M.; Horsky, Dan; Narasimhan, Chakravarthi
March 1988
Marketing Science;Spring88, Vol. 7 Issue 2, p169
Academic Journal
The sales-marketing mix relationships for brands of a product class are often modelled as a multiple equation system. Whenever sales are expressed as shares, such systems are sum-constrained and therefore singular. Singular systems can be calibrated by deleting one equation from the model and estimating the remaining equations using the seemingly unrelated regressions (SUR) technique. A two-step procedure is typically adopted whenever the disturbance covariance structure is unknown. This study highlights an issue often overlooked by marketing researchers. Traditional two-step estimators that delete one equation at the initial step usually lack invariance to the equation deleted. Invariant estimates are available using maximum likelihood estimation (MLE) or iterated least squares. However, relative to simple, two-step procedures, these approaches require more computation and are often less accurate in small samples. We describe several ways to obtain invariant, two-step estimators and conduct a sampling experiment to compare the finite sample behavior of these and other estimators to an iterated solution. Our results show that a "balanced" two-step estimator which deletes no equations initially outperforms the iterated solution over a wide range of conditions. Our results also show that a constrained, single-step estimator which deletes no equations from the system frequently outperforms all two-step methods examined. Further, the results of our study also apply to nonlinear specifications of market share models since the invariance issue arises in this context as well.


Related Articles

  • NUMERICAL DYNAMIC MODELING AND DATA DRIVEN CONTROL VIA LEAST SQUARE TECHNIQUES AND HEBBIAN LEARNING ALGORITHM. Xiaodong Liu; Yanli Zhang // International Journal of Numerical Analysis & Modeling;2010, Vol. 7 Issue 1, p66 

    The modelling and controlling for complex dynamic systems which are too complicated to establish conventionally mathematical mechanism models require new methodology that can utilize the existing knowledge, human experience and historical data. Fuzzy cognitive maps (FCMs) are a very convenient,...

  • Multicollinearity in Structural Equation Models With Unobservable Variables. Jagpal, Harsharanjeet S. // Journal of Marketing Research (JMR);Nov82, Vol. 19 Issue 4, p431 

    The author develops a new ridge estimator for the treatment of multicollinearity in structural equation models with unobservable variables. The method is illustrated by a simple model of advertising in the multiproduct firm. For the nonlinear interaction model, ordinary partial least squares...

  • Share-Ratio Estimation of the Nested Multinomial Logit Model. BECHTEL, GORDON G. // Journal of Marketing Research (JMR);May90, Vol. 27 Issue 2, p232 

    Recently the nested multinomial logit (NMNL) model has been proposed to remedy the "independence of irrelevant alternatives" (IIA) property of the classical multinomial logit (MNL) model. The author brings the NMNL model under generalized least squares theory, thereby enhancing its estimation...

  • CONTRASTING TOTAL LEAST SQUARES WITH ORDINARY LEAST SQUARES PART I: BASIC IDEAS AND RESULTS. Scariano, Stephen M.; Barnett, II, William // Mathematics & Computer Education;Spring2003, Vol. 37 Issue 2, p141 

    The article presents the first part of a two-article series which contrasts the ordinary least squares and total least squares method. It stated that the method of total least squares is a reasonable competitor that can be exploited for classroom presentation. It studies the basic criteria...

  • Pooling Issues and Methods in Regression Analysis: Some Further Reflections. Bass, Frank M.; Wittink, Dick R. // Journal of Marketing Research (JMR);May78, Vol. 15 Issue 2, p277 

    In this article the authors address commentary made on their paper which examined pooling issues and methods. They note that the main argument made against their paper was that they should have employed Wallace's test to determine whether pooling would be appropriate using weaker criteria than...

  • Estimating Demand Response with Panel Data. Lecocq, S├ębastien; Robin, Jean-Marc // Empirical Economics;Nov2006, Vol. 31 Issue 4, p1043 

    In this paper, we extend to panel data the iterated linear least squares estimator of Blundell and Robin (in J Appl Econometrics 14: 209-232 1999). It is shown to be consistent when total expenditure and regression residuals are correlated, either because of simultaneity or because of unobserved...

  • DERIVING THE GLS TRANSFORMATION PARAMETER IN ELEMENTARY PANEL DATA MODELS. Jefferson, Philip N. // American Economist;Spring2005, Vol. 49 Issue 1, p45 

    The Generalized Least Squares (GLS) transformation that eliminates serial correlation in the error terms is central to a complete understanding of the relationship between the pooled OLS, random effects, and fixed effects estimators. A significant hurdle to attainment of that understanding is...

  • R2 VERSUS r2. Shu-Ping Hu // Journal of Cost Analysis & Parametrics (Society of Cost Estimati;Fall2010, Vol. 3 Issue 2, p13 

    Cost estimating relationships (CERs) with multiplicative-error assumptions are commonly used in cost analysis. Consequently, we need to apply appropriate statistical measures to evaluate a CER's quality when developing multiplicative error CERs such as minimum-unbiased-percentage error (MUPE)...

  • Least Squares Symmetrizable Solutions for a Class of Matrix Equations. Fanliang Li // Applied Mathematics;May2013, Vol. 4 Issue 5, p741 

    In this paper, we discuss least squares symmetrizable solutions of matrix equations (AX = B, XC = D) and its optimal approximation solution. With the matrix row stacking, Kronecker product and special relations between two linear subspaces are topological isomorphism, and we derive the general...


Read the Article


Sorry, but this item is not currently available from your library.

Try another library?
Sign out of this library

Other Topics