Stanley R. Johnson

Advanced econometric methods

1 Introduction.- 2 Review of Ordinary Least Squares and Generalized Least Squares.- 3 Point Estimation and Tests of Hypotheses in Small Samples.- 4 Large Sample Point Estimation and Tests of Hypotheses.- 5 Stochastic Regressors.- 6 Use of Prior Information.- 7 Preliminary Test and Stein-Rule Estimators.- 8 Feasible Generalized Least Squares Estimation.- 9 Heteroscedasticity.- 10 Autocorrelation.- 11 Lagged Dependent Variables and Autcorrelation.- 12 Unobservable Variables.- 13 Multicollinearity.- 14 Varying Coefficient Models.- 15 Models That Combine Time-Series and Cross-Section Data.- 16 The Analysis of Models with Qualitative or Censored Dependent Variables.- 17 Distributed Lags.- 18 Uncertainty in Model Specification and Selection.- 19 Introduction to Simultaneous Equations Models.- 20 Identification.- 21 Limited Information Estimation.- 22 Full Information Estimation.- 23 Reduced Form Estimation and Prediction in Simultaneous Equations Models.- 24 Properties of Dynamic Simultaneous Equations Models.- 25 Special Topics in Simultaneous Equations.- Appendix Estimation and Inference in Nonlinear Statistical Models.- A.1 Nonlinear Optimization.- A.1.1 Method of Steepest Ascent.- A.1.2 The Method of Newton.- A.1.3 Method of Quadratic Hill Climbing.- A.1.4 Numerical Differentiation.- A.2 Maximum Likelihood Estimation.- A.2.1 Use of the Method of Newton.- A.2.2 Method of Scoring.- A.2.3 The Method of Berndt, Hall, Hall, and Hausman.- A.2.4 Asymptotic Tests Based on the Maximum Likelihood Method.- A.2.4a The Wald Test.- A.2.4b The Lagrange-Multiplier Test.- A.2.4c The Likelihood Ratio Test Statistic.- A.2.4d Concluding Remarks.- A.3 Nonlinear Regression.- A.4 Summary and Guide to Further Readings.- A.5 References.

A General Model of Rent Seeking for Public Goods

This paper generalizes the model of collective rent-seeking over a public good. Expanding the rent seekers consumption bundle to include preferences over the public good and a private good, our results suggest collective rent-seeking is positively related to group size. Although free riding exists within a group, there is not a one-for-one tradeoff. In addition, rent seeking increases with wealth. Finally, total effort expended by both groups increases if either group increases in membership size, except in the case of an extremely lopsided contest. The key condition underlying these results is that the marginal utility of the public good is not inversely related to the private good.

Nutrient Availability, Food Costs, and Food Stamps

Nutrient consumption patterns of low income households were analyzed using data from the special low income component of the 1977–78 Nationwide Food Consumption Survey. A system of simultaneous equations for estimating food cost and diet component availability levels in home food supply was applied. The derived reduced form showed that Food Stamp Program participation had a positive impact on diet component availability levels. Impacts of other socioeconomic variables examined for diet component availability were generally larger than those for program participation and income. Diet component availability levels were relatively constant across households with different income and Food Stamp Program bonus levels.

Uncertainty and Split Nitrogen Application in Corn Production

Investigates the split application of nitrogen for rainfed corn in the Midwest using two-stage decision model. Ex post analysis of late nitrogen application; Ex ante analysis of preplant nitrogen application: Use of information from the Erosion Productivity Impact Calculator (EPIC); Implications for regulating nitrogen use; Empirical results for Nashua, Iowa nitrogen application.

An Hedonic Index for Breakfast Cereals

The hedonic index technique is used to estimate the implicit prices for units of breakfast cereal characteristics. Analysis showed that buyers value many attributes; the value placed on each is dependent upon the class of cereal evaluated. Results suggest this technique has potential for assessing buyers purchasing behavior for nondurable goods.

Maximum likelihood estimation of linear expenditure systems with serially correlated errors: An application

Abstract The present paper uses Canadian data for the period 1947–1972 and three commodity groups to examine the empirical importance of restrictions imposed by autocorrelated disturbances on the static linear expenditure system, LES. For comparison a simple habit persistence model is also estimated. Results of applications of likelihood ratio tests indicate that autocorrelation is present in the data, that a simple habit persistence hypothesis on the structure is implied and that the restrictions imposed by the form of the utility function and maximization problem are inconsistent with the data whether or not adjusted for the autocorrelation in errors. Despite these differences, the estimated price and income elasticities remain fairly constant across the various specifications of the LES that were considered.

Regional econometric models that reflect labor market relations

Abstract Regional economic forecasting is often hampered by a lack of reliable data. This study attempts to improve the efficiency of forecasts by incorporating interindustry linkages in wage and employment determination into an econometric model. Prior information of historical linkages is utilized in the form of restricted least squares and applied to the Louisville metropolitan area. We provide a step toward the specification of reliable models that reflect the local institutional framework which is often ignored in neoclassical, labor market-based model specifications. Comparisons on forecast accuracy with other structural models found in the literature favor the use of local institutional factors, though univariate time series models still tend to outperform the econometric approach.

Children's frequency of eating, total sugar intake and weight/height stature

Abstract A nationwide survey of 972 individuals ages 5 through 18 years was utilized to assess if a relationship existed between eating patterns, total sugar intake and weight/height ratios. Cross-tabulations and regression analysis were employed to test the hypothesized relations. Results indicated that there existed no general relationship between eating patterns and occurrence of obesity. The authors suggest additional analyses utilizing more homogeneous samples be undertaken before conclusions are formed concerning relationships of snack consumption and weight/height ratios.

A Comparative Analysis of State Regulations for Use of Agricultural Chemicals

Policies to regulate pesticides at the national level have not changed as new issues, such as groundwater contamination, have emerged. Therefore, various states are responding. This paper discusses recent state initiatives in regulating pesticides to prevent groundwater contamination, and suggests trends in these policies. A survey of legislators who have sponsored pesticide/groundwater contamination legislation is presented. Survey results show the factors affecting bill introduction, the influence of interest groups on the bills, important issues in bill debate, and trends in future legislation. A statistical analysis of the survey states and survey responses is also presented, using economic, political, and physical factors as explanatory variables.

Compromise solution for economic-environmental decisions in agriculture*

Least-cost production versus the environmental on- and off-site erosion damage of agriculture is evaluated in a policy context for a major Corn Belt watershed. Compromise programming, previously utilized in firm-level multi-criteria decision-making problems, is applied to a regional agricultural production model with environmental policy goal trade-offs. The crop sector model allocates land, water, labor, capital and commodity-program base acres to crop production. Production options include four conservation practices, three tillage methods and several crop rotations. Crop yield and fertilizer levels are dependent upon erosion. Cropping options selected allow for both wind and water erosion. The vector of objectives include three minimization functions: current production cost, future value of productivity loss and sediment damage. Vector optimization technique was used to generate the payoff matrix containing efficient but simultaneously unobtainable solutions. Given the ideal but infeasible solution vector we generated efficient solutions in the compromise subset corresponding to the L1, L2 and L∞ metrics. Trade-off relations were developed using the non-inferior set estimation technique.