Richard D. Green

Elasticities in AIDS Models

In the literature, a variety of approaches have been used to calculate demand elasticities in almost ideal demand system (AIDS) models of demand. It is common to estimate the linear approximate almost ideal demand system (LA/AIDS) instead of the AIDS. When the LA/AIDS is estimated, all of the previously reported approaches to compute elasticities are theoretically incorrect. This paper presents correct formulas for LA/AIDS elasticities and illustrates the potential errors from using incorrect computing formulas.

Maintaining and Testing Separability in Demand Systems

We derive a general elasticity representation of the necessary and sufficient conditions for direct weak separability of the utility function. Parametric restrictions required to implement the separability conditions are presented for three common demand systems: the Almost Ideal, Translog, and Rotterdam. Our empirical application uses the Rotterdam model to test a few separable structures within a complete U.S. demand system emphasizing food commodities. Results, based on size-corrected likelihood ratio tests, provide support for commonly used separability assumptions about food and meat demand.

Estimating Elasticities with the Linear Approximate Almost Ideal Demand System: Some Monte Carlo Results

The almost ideal demand system (AIDS) model has many desirable theoretical properties but usually it is estimated using a linear approximation. The quality of the approximation to the true AIDS depends on the parameters and the collinearity among the exogenous price variables. In the literature, four alternative formulas have been used to compute elasticities of a demand system that is assumed to be of the AIDS form using parameters estimated in the linear approximate AIDS. Month Carlo experiments indicate that two of those four alternatives, as typically applied, are highly inaccurate but the other two are quite accurate. Copyright 1994 by MIT Press.

Whither Armington Trade Models

The Armington trade model distinguishes commodities by country of origin, and import demand is determined in a separable two-step procedure. This framework has been applied to numerous international agricultural markets with the objective of modeling import demand. In addition, computable general equilibrium (CGE) models commonly employ the Armington formulation in the trade linkage equations. The purpose of this paper is to test the Armington assumptions of homotheticity and separability with data from the international cotton and wheat markets. Both parametric and nonparametric tests were performed, and the empirical results reject the Armington assumptions. This has important implications for international trade modeling and CGE modeling.

Statistical methods for bankruptcy forecasting

Abstract The three statistical models most often used for bankruptcy forecasting are multiple discriminant analysis, the linear probability model, and logistic regression. This paper compares and contrasts these models, evaluates their effectiveness and statistical properties, and considers how well each model fits with a theory of financial distress.

Elasticities in AIDS Models: A Clarification and Extension

expenditure shares (i.e., InP = Ij wj InP,). A common approach is to treat expenditure shares as constant parameters in the Stones index when taking derivatives for elasticities. We developed corrected formulas for uncompensated price elasticities by using derivatives that took into account the effects of price changes on the shares in the price index. In the process, and as a side-issue, we also noted (p. 443) that the differences in uncompensated elasticities in the literature (ri)

Alternative Estimates of Fed Beef Supply Response to Risk

Supply response models for fed beef incorporating risk by including both the mean and variance of output price are developed, estimated, and compared. Six different estimates of the mean and variance are obtained from futures prices; ARIMA processes; and naive, adaptive, and rational expectations models. Empirical estimates are compared using nonnested testing procedures and indicate that the choice of the expectations model significantly influences elasticity estimates and whether supply response is positive or negative. Empirical evidence does not support any one model in particular, suggesting that expectations are heterogenous rather than homogenous.

Suburban growth controls and the price of new housing

Abstract The effect of Petaluma, Californias growth control program on new house prices is examined by a comparative (quasi-experimental) design, using statistical controls. The difference in price increases between Petaluma and two comparison cities is calculated for houses of standard quality and for average actual houses in each city. It was found that prices of standard quality houses increased significantly more in Petaluma than in one of the comparison cities but not the other. A similar pattern was observed for actual houses. Reasons for the observed results and specific causes of the growth control effect are analyzed.

Behavioral Changes of Chronic Schizophrenic Patients Given L-5-Hydroxytryptophan

Oral administration of the serotonin precursor L-5-hydroxytryptophan with a peripheral decarboxylase inhibitor produced Mild to moderate improvement in six of seven chronic undifferentiated schizophrenic patients who were resistant to phenothiazine treatment, as compared to an oral administration of a placebo. Two of four chronic paranoid schizophrenic patients who were resistant to phenothiazine treatment became worse with 5-hydroxytryptophan, one improved. It is presumed that these psychological changes were directly or indirectly produced from increases in brain serotonin. Indirect data from animals and humans indicate that there may be an abnormality in serotonin metabolism in some schizophrenics. While our data are consistent with this hypothesis, other explanations for our data must be entertained.

Standard Errors for Elasticities: A Comparison of Bootstrap and Asymptotic Standard Errors

This article compares two methods of deriving standard errors for elasticities in a linear expenditure demand system with first-order autoregressive errors. The first is the ordinary Taylor series method and the second is Efrons bootstrap. In an example problem, these two methods yielded similar values for the standard errors, with the exception of the income elasticity standard errors, for which the asymptotic standard errors were apparently too large by a factor of two.