Abstract: A previous study showed that imposing economic restrictions improves the forecasting ability of food demand systems, thus warranting their use even when they are rejected in-sample. This article evaluates whether this result is due to economic restrictions enhancing degrees of freedom or containing nonsample information. Results indicate that restrictions improve forecasting ability even when they are not derived from economic theory, but theoretical restrictions forecast best.
Abstract: We compare the ability of three preference elicitation methods (hypothetical choices, nonhypothetical choices, and nonhypothetical rankings) and three discrete-choice econometric models (the multinomial logit [MNL], the independent availability logit [IAL], and the random parameter logit [RPL]) to predict actual retail shopping behavior in three different product categories (ground beef, wheat flour, and dishwashing liquid). Overall, we find a high level of external validity. Our specific results suggest that the nonhypothetical elicitation approaches, especially the nonhypothetical ranking method, outperformed the hypothetical choice experiment in predicting retail sales. We also find that the RPL can have superior predictive performance, but that the MNL predicts equally well in some circumstances. experiment.
Abstract: We compare the ability of three preference elicitation methods (hypothetical choices, non-hypothetical choices, and non-hypothetical rankings) and three discrete-choice econometric models (the multinomial logit, the independent availability logit, and the random parameter logit) to predict actual retail shopping behavior in three different product categories (ground beef, wheat flour, and dishwashing liquid). Overall, across all methods, we find a reasonably high level of external validity. Our results suggest that the non-hypothetical elicitation approaches, especially the non-hypothetical ranking, outperformed the hypothetical choice experiment in predicting retail sales. We also find that the random parameter logit can have superior predictive performance, but that the multinomial logit predicts equally well in some circumstances.
Abstract: In this work, we analyse EU soybean and maize imports using a demand system borrowed from the differential approach to firm theory. Alongside providing own-price and cross-price (i. e. cross-country) elasticities for these two products, we test whether source-specific characteristics exert any influence on complementarity and substitution patterns between international exporters. Specifically, we look at country differences stemming from supply chain efficiency and the asynchronous approval of Genetically Modified (GM) varieties. We do so by introducing two measurements for such features into a linear demand model specified by Laitinen and Theil (1978). Estimation results suggest that the EU import structure is not affected by differences in supply chain efficiency between overseas suppliers while, depending on the product, asynchronous approval does seem to have an influence. We find that imports of maize are more sensitive than those of soybeans to differences in approval statuses between international exporters and the EU. Since soybean availability is a limiting factor for the EU feed industry, avoiding stock shortages may be a priority for European importers, hence the weaker effect of asynchronous approval. On the other hand, the substantial EU self-sufficiency for maize places more emphasis on product characteristics and prices.
Abstract: Given the importance of EU demand for chilled fish fillets to the exporting sectors in Tanzania and Uganda, this study estimated the EUs import demand for fillets by country of origin to assess the competitiveness of exporters. Results imply that prices in Tanzania and Uganda had an insignificant impact on total imports expenditures in the EU. Conditional and unconditional cross-price effects indicated that exports from Lake Victoria did not compete with exports from other suppliers, such as Iceland, Norway and ROW. Import demand forecasts showed that market share in the EU should remain relatively unchanged given the trend in prices.
Abstract: Ever since Richard Stone (1954) first estimated a system of demand equations derived explicitly from consumer theory, there has been a continuing search for alternative specifications and functional forms. Many models have been proposed, but perhaps the most important in current use, apart from the original linear expenditure system, are the Rotterdam model (see Henri Theil, 1965, 1976; Anton Barten) and the translog model (see Laurits Christensen, Dale Jorgenson, and Lawrence Lau; Jorgenson and Lau). Both of these models have been extensively estimated and have, in addition, been used to test the homogeneity and symmetry restrictions of demand theory. In this paper, we propose and estimate a new model which is of comparable generality to the Rotterdam and translog models but which has considerable advantages over both. Our model, which we call the Almost Ideal Demand System (AIDS), gives an arbitrary first-order approximation to any demand system; it satisfies the axioms of choice exactly; it aggregates perfectly over consumers without invoking parallel linear Engel curves; it has a functional form which is consistent with known household-budget data; it is simple to estimate, largely avoiding the need for non-linear estimation; and it can be used to test the restrictions of homogeneity and symmetry through linear restrictions on fixed parameters. Although many of these desirable properties are possessed by one or other of the Rotterdam or translog models, neither possesses all of them simultaneously. In Section I of the paper, we discuss the theoretical specification of the AIDS and justify the claims in the previous paragraph. In Section II, the model is estimated on postwar British data and we use our results to test the homogeneity and symmetry restrictions. Our results are consistent with earlier findings in that both sets of restrictions are decisively rejected. We also find that imposition of homogeneity generates positive serial correlation in the errors of those equations which reject the restrictions most strongly; this suggests that the now standard rejection of homogeneity in demand analysis may be due to insufficient attention to the dynamic aspects of consumer behavior. Finally, in Section III, we offer a summary and conclusions. We believe that the results of this paper suggest that the AIDS is to be recommended as a vehicle for testing, extending, and improving conventional demand analysis. This does not imply that the system, particularly in its simple static form, is to be regarded as a fully satisfactory explanation of consumers' behavior. Indeed, by proposing a demand system which is superior to its predecessors, we hope to be able to reveal more clearly the problems and potential solutions associated with the usual approach.
Abstract: Some decision rules for discriminating among alternative regression models are proposed and mutually compared. They are essentially based on the Akaike Information Criterion as well as the Kullback-Leibler Information Criterion (KLIC) : namely, the distance between a postulated model and the true unknown structure is measured by the KLIC. The proposed criteria combine the parsimony of parameters with the goodness of fit. Their relationships with conventional criteria are discussed in terms of a new concept of unbiasedness .
Abstract: This paper is concerned with testing for causation, using the Granger definition, in a bivariate time-series context. It is argued that a sound and natural approach to such tests must rely primarily on the out-of-sample forecasting performance of models relating the original (non-prewhitened) series of interest. A specific technique of this sort is presented and employed to investigate the relation between aggregate advertising and aggregate consumption spending. The null hypothesis that advertising does not cause consumption cannot be rejected, but some evidence suggesting that consumption may cause advertising is presented.
Abstract: When a regression problem contains many predictor variables, it is rarely wise to try to fit the data by means of a least squares regression on all of the predictor variables. Usually, a regression equation based on a few variables will be more accurate and certainly simpler. There are various methods for picking “good” subsets of variables, and programs that do such procedures are part of every widely used statistical package. The most common methods are based on stepwise addition or deletion of variables and on “best subsets.” The latter refers to a search method that, given the number of variables to be in the equation (say, five), locates that regression equation based on five variables that has the lowest residual sum of squares among all five variable equations. All of these procedures generate a sequence of regression equations, the first based on one variable, the next based on two variables, and so on. Each member of this sequence is called a submodel and the number of variables in the e...