Consistent estimation in the bilinear multivariate errors-in-variables model

A bilinear multivariate errors-in-variables model is considered. It corresponds to an overdetermined set of linear equations AXB=C, A∈ℝ^m×n, B∈ℝ^p×q, in which the data A, B, C are perturbed by errors. The total least squares estimator is inconsistent in this case.  An adjusted least squares estimator is constructed, which converges to the true value X, as m →∞, q →∞. A small sample modification of the estimator is presented, which is more stable for small m and q and is asymptotically equivalent to the adjusted least squares estimator. The theoretical results are confirmed by a simulation study. Acknowledgements. We thank two anonymous reviewers for their suggestions and corrections.? A. Kukush is supported by a postdoctoral research fellowship of the Belgian office for Scientific, Technical and Cultural Affairs, promoting Scientific and Technical Collaboration with Central and Eastern Europe.? S. Van Huffel is a full professor with the Katholieke Universiteit Leuven.? I. Markovsky is a research assistant with the Katholieke Universiteit Leuven.? This paper presents research results of the Belgian Programme on Interuniversity Poles of Attraction (IUAP V-22), initiated by the Belgian State, Prime Minister's Office – Federal Office for Scientific, Technical and Cultural Affairs of the Concerted Research Action (GOA) projects of the Flemish Government MEFISTO-666 (Mathematical Engineering for Information and Communication Systems Technology), of the IDO/99/03 project (K.U. Leuven) “Predictive computer models for medical classification problems using patient data and expert knowledge”, of the FWO projects G.0078.01, G.0200.00, and G0.0270.02.? The scientific responsibility is assumed by its authors.     

Multiple Imputation in Multivariate Problems When the Imputation and Analysis Models Differ

Bayesian multiple imputation (MI) has become a highly useful paradigm for handling missing values in many settings. In this paper, I compare Bayesian MI with other methods – maximum likelihood, in particular—and point out some of its unique features. One key aspect of MI, the separation of the imputation phase from the analysis phase, can be advantageous in settings where the models underlying the two phases do not agree.     

A theory of global strategic alliances: The case of the global auto industry

The theory articulated in this paper suggests that the desire to reduce demand and competitive uncertainty are two separate, important motives for alliance formation. Taking this as a starting point, we predict the configuration of horizontal alliances that we might expect to observe within an industry when firms experience these uncertainties to different degrees. An empirical test of this theory using data from the global auto industry yields results consistent with the view (1) that alliances are a device for reducing both the uncertainties that arise from unpredictable demand conditions and those that arise from competitive interdependence, and (2) that variation of demand uncertainty and competitive uncertainty across firms explains differentials in both the intensity and structure of their horizontal alliance activity.     

Estimating the effectiveness of extension information systems using farm trials and subjective probabilities

A method of combining survey data and Dairy Herd Improvement Association (DHIA) records to achieve low cost farm trials is presented. Farm trials and surveys of current practices and production responses are useful to identify yield gaps between expectations predicted from experimental findings and actual field results. Different management schemes can be ranked using subjective probabilities and stochastic dominance to enhance successful implementation of research findings and to increase the feedback between researchers, extension workers, and producers. A survey of current masitis control practices and expected milk yield response is the example. Combining survey results with DHIA records allowed estimating the relationship between somatic cell counts (SCC) and milk yield. Eliciting beliefs about the relationship between SCC and milk yield showed that producers agreed with predictions from the statistical model. Subjective probabilities about SCC and mastitis control practices showed that our sample of experts and producers consistently ranked the different practices but extension agents had no consensus about the the most or least effective ones.     

Problem shifts and market research: The role of networks in business relationships

This article discusses changes in approach to the study of business relationships, and sees the paradigm concept as constituting an inadequate basis for describing gradually evolving changes in the perception of business. The frequent references to paradigms appear to focus more on state aspects than on change aspects of theory building. The article attempts to trace some of the changes in the perception of markets preceding the emergence of network theories, and relates the presence of an interactional approach to business studies to the concept of problem shifts (Lakatos). The development of interactional approaches in marketing is seen as representing a shift away from theory building commonly found in economics and organizational theory. Changes in the approach to the study of business relationships are interpreted as theoretical or empirical problem shifts, and are thus related to the a priori and a posteriori relevance of the continual evolution of alternative business theories. The article explores some of the epistemological relevance of the network theories to the study of market relationships, and the potential impact on continual changes in approach to the study of markets.     

A comparison of several approximations to the power function of the likelihood ratio test in covariance structure analysis*

In this paper we compare alternative asymptotic approximations to the power of the likelihood ratio test used in covariance structure analysis for testing the fit of a model. Alternative expressions for the noncentrality parameter (ncp) lead to different approximations to the power function. It appears that for alternative covariance matrices close to the null hypothesis, the alternative ncp's lead to similar values, while for alternative covariance matrices far from H_o the different expressions for the ncp can conflict substantively. Monte Carlo evidence shows that the ncp proposed in Satorra and Saris (1985) gives the most accurate power approximations.