Steven T. Yen

Two-Step Estimation of a Censored System of Equations

A consistent two-step estimation procedure is proposed for a system of equations with limited dependent variables. Monte Carlo simulation results suggest the procedure outperforms an existing two-step method. Copyright 1999, Oxford University Press.

Quasi- and Simulated-Likelihood Approaches to Censored Demand Systems: Food Consumption by Food Stamp Recipients in the United States

A quasi-maximum-likelihood estimator is proposed and applied to a censored Translog demand system for foods, using a sample of food stamp recipients in the United States. The procedure produces remarkably close parameter and elasticity estimates to those of the simulated-maximum-likelihood procedure. A two-step procedure is also considered but it produces different elasticities. Demands are found to be price elastic for pork and fish but price inelastic for all other food products. Gross complementarity and net substitutability are obvious but these cross-price effects are much less pronounced than own-price and total food expenditure effects. Copyright 2003, Oxford University Press.

Food Stamp Program Participation and Food Insecurity: An Instrumental Variables Approach

The relationship between Food Stamp Program (FSP) participation and household food insecurity (FI) is investigated using data from the 1996–1997 National Food Stamp Program Survey. Endogeneity of FSP participation is accommodated with an instrumental variables approach. In contrast to other findings reported in the literature, results suggest participation in the FSP reduces the severity of FI. Sociodemographic variables play important roles in FSP participation and FI. Underreporting of FSP participation and limited observations of food-insecure households in previous studies may have also been factors. Copyright 2008, Oxford University Press.

A Multivariate Sample-Selection Model: Estimating Cigarette and Alcohol Demands with Zero Observations

Heckmans sample-selection model is generalized to the multivariate case and used to estimate demands for cigarettes, beer, and wine by individuals in the United States. The specification allows correlations between the error terms of multiple selection and level equations, and nests the popular bivariate sample-selection model and two-part model. Empirical results suggest that the proposed model performs better than the restricted specifications. Differentiated effects of variables on probabilities and levels of consumption also suggest rejection of the Tobit system. Gender differences are present, and demographic variables are more important than income in determining consumption of cigarettes, beer, and wine.

Household demand for fats and oils: two-step estimation of a censored demand system

A censored system of household fat and oil demand equations is estimated with a two-step procedure, using cross-sectional data from the 1987–1988 US Nationwide Food Consumption Survey. Own price and total expenditure elasticities are close to unity and there is no evidence of gross substitutability. Compensated elasticities suggest net substitution among the products considered.

A Sample Selection Approach to Censored Demand Systems

The multivariate sample selection model is extended to a nonlinear equation system with partial selection and applied to household meat consumption in China. Elasticity estimates differ from those obtained from conventional maximum likelihood and Tobit estimates. Chinese meat products are gross complements while net substitution also exists in some cases.

Household Consumption of Cheese: An Inverse Hyperbolic Sine Double-Hurdle Model with Dependent Errors

The dependent double-hurdle model is generalized by an inverse hyperbolic sine transformation of the dependent variable. The resulting specification features a flexible parameterization, accommodates heteroskedastic errors, and nests a range of common limited dependent variable models. Results for U.S. household cheese consumption suggest that the homoskedastic and normal double-hurdle model is misspecified. Income elasticities are small and vary across household groups. Foodstamp recipients are more responsive to income changes than nonrecipients. Foodstamp recipients are also less likely to consume cheese but, conditional on consumption, spend more than nonrecipients. Copyright 1997, Oxford University Press.

A censored system of cigarette and alcohol consumption

Many previous studies of US cigarette and alcohol consumption have focused on single equations. However, the addictive nature of consumption suggests that it is more appropriate to model these products as a system. We propose a two-step estimation procedure and apply the procedure to a system of equations for cigarette, beer and wine consumption. Results differ from those derived from an existing two-step procedure. Findings suggest that consumption of cigarettes, beer and wine is responsive to income changes but the elasticities are rather small. Personal physiques, education, age, race, ethnicity, health, gender, employment status, and regions also play significant roles in consumption.

An econometric analysis of household donations in the USA

This study investigates philanthropy, an American tradition. A censored system of donation equations is estimated by full-information maximum likelihood, using data from the 1995 Consumer Expenditure Survey. Results suggest that the censored system estimates are more appropriate than the single-equation estimates attempted in much of the donation literature. Income, age, and education are contributing factors of donation, regardless of whether it is to charity, religious organizations, or other organizations.

Individual cigarette consumption and addiction: A flexible limited dependent variable approach

A variety of limited dependent variable models have been used in microeconometric studies of smoking. We present a general model of the simultaneous decisions of how much to smoke and whether to quit, which allows a more flexible parameterization and distributional assumptions. The model incorporates the fixed costs associated with quitting and allows for the separate influence of addiction on participation and consumption. The model is applied to data on smoking from the 1984-85 UK Health and Lifestyle Survey, and is shown to outperform a range of nested models that have been used extensively in the empirical literature.