ESTIMATING SUPPLY DESPITE A QUOTA: BRAZIL'S SUPPLY OF SUGAR

In this paper a model is developed from which the supply curve, of a stochastically produced good, can be estimated when the observations are constrained by a quota. It is shown that, for any (price, quantity) combination observed in the presence of a quota, we can simulate a corresponding (price, quantity) combination which could have been observed in the absence of a quota. This procedure is then employed to construct "free-market" observations on Brazil's supply of sugar, with which the supply curve is then estimated, and production in the absence of sugar quotas inferred.     

Distribution of power and agrarian reform: An explanatory model applied to India and Chile

This article appraises within an interdisciplinary framework key social, political, and economic aspects of the distribution of power in agrarian societies, and their interaction with various agrarian reform measures. Reform measures investigated include land reform proper, tenancy and wage regulations, land taxes, institutional credit, and productivity promotion measures.Use is made of alternate models that simulate economic development and agrarian reform under alternative sociopolitical regimes. The models distinguish four actor-groups: landlords, peasants, nonagriculturalists, and the state. Regimes differ, on the one hand, according to which actor-group is in power, and, on the other hand, according to which instruments of economic policy the state employs. The models are empirically applied and used to analyze the impact of agrarian reform measures in the 1960s and 1970s in two different regimes, India and Chile (under Allende).     

Father loss and father absence in preschool children

The role of the father in child development is discussed. Children's reactions to the death, separation, or divorce of their fathers are explored. Implications for diagnosis are stated along with specific recommendations for treatment. Crucial factors in diabnosis and treatment were concluded to be: the developmental stage of the child at the time of the loss, the reason for the father's absence, the state of the relationship with the father prior to the loss, the mother's reaction to the loss of her husband, and the environment provided for the child subsequent to the loss.     

AN APPROXIMATE MAXIMUM LIKELIHOOD ESTIMATOR FOR CHAIN BINOMIAL MODELS

Using the Poisson approximation to the Binomial distribution, we construct an approximate maximum likelihood estimator (MLE) for a class of chain binomial models. Our estimator proves to have properties which may make it preferable to the exact WLE.     

Changes in Americans' Policy Preferences, 1935-1979

Using responses to 3,315 survey questions asked of nationalsamples, we examine how policy preferences of Americans havechanged over the last 45 years. The data indicate that therehas been considerable stability in public opinion: responsesto half the 613 repeated policy items show no significant changeat all; approximately half the detectable changes were lessthan 10 percentage points; and rarely did preferences fluctuatesignificantly back and forth within a short time period. Foreignpolicy changes were no larger or more frequent than domestic,but they did tend to occur more abruptly. When opinion shiftsoccurred, they were not random or capricious; they were usuallyrelated to important changes in citizens' social and economicenvironments. Rapid shifts generally coincided with major eventsin international affairs or the economy.     

The negotiating style of Saudi industrial buyers: An empirical investigation

This study investigates the negotiating styles of Saudi industrial buyers. Thomas’s Topology as measured by Rahim’s instrument is used in this study. The results suggest that three are two dominant styles of negotiation in the Saudi context, namely: the competitive and collaborative styles.     

Telephone survey respondents' reactions to questions regarding interpersonal violence

Concerns have been raised regarding the appropriateness of asking about violence victimization in telephone interviews and whether asking such questions increases respondents' distress or risk for harm. However, no large-scale studies have evaluated the impact of asking such questions during a telephone interview. This study explored respondents' reactions to questions regarding violence in two large recently completed telephone surveys. After respondents were asked about violence, they were asked if they thought surveys should ask such questions and whether they felt upset or afraid because of the questions. In both surveys, the majority of respondents (regardless of their victimization history) were willing to answer questions about violence and were not upset or afraid because of the questions. More than 92% of respondents thought such questions should be asked. These results challenge commonly held beliefs and assumptions and provide some assurance to those concerned with the ethical collection of data on violent victimization.     

Nonparametric empirical Bayes for the Dirichlet process mixture model

The Dirichlet process prior allows flexible nonparametric mixture modeling. The number of mixture components is not specified in advance and can grow as new data arrive. However, analyses based on the Dirichlet process prior are sensitive to the choice of the parameters, including an infinite-dimensional distributional parameter G ₀. Most previous applications have either fixed G ₀ as a member of a parametric family or treated G ₀ in a Bayesian fashion, using parametric prior specifications. In contrast, we have developed an adaptive nonparametric method for constructing smooth estimates of G ₀. We combine this method with a technique for estimating α, the other Dirichlet process parameter, that is inspired by an existing characterization of its maximum-likelihood estimator. Together, these estimation procedures yield a flexible empirical Bayes treatment of Dirichlet process mixtures. Such a treatment is useful in situations where smooth point estimates of G ₀ are of intrinsic interest, or where the structure of G ₀ cannot be conveniently modeled with the usual parametric prior families. Analysis of simulated and real-world datasets illustrates the robustness of this approach.