Panel data regression for counts

A new panel data model for count data is introduced. We suggest alternative estimators, such as pseudo maximum likelihood and generalized method of moments, of structural and nuisance parameters. In addition, different test statistics of independence and overdispersion are obtained. The small sample performance of the estimators and tests are evaluated in Monte Carlo experiments. The model is applied to the number of days absent in Sweden 1981–1991 for a panel of Swedish male workers.     

Interrelationships between fertility and marital dissolution: Results of a simultaneous logit model

Previous studies have failed to examine (a) the simultaneity of decisions to bear children and to dissolve a marriage and (b) the effects of the childbearing in the period just before dissolution on the likelihood of disruption. We attempted to resolve both problems by developing a simultaneous logit model of the interrelationship between the probability of separation and of having a birth during this period (when dissolution presumably is being considered). The model was estimated at successive durations of first marriage, using data for white women in the 1970 National Fertility Survey. The results indicate that childbearing patterns— number of children and age of youngest child at the beginning of the marital interval being studied and fertility during the interval—did not influence the likelihood of separation in simple or consistent ways over the marital life course, nor did marital strife (as indicated by separation) seem to affect childbearing throughout marriage.     

Family violence: A microeconomic approach

A model of violence between adult family members is developed by integrating material from the sociological theories of family violence and social exchange, and the economic theories of crime and the family. Based on this model a decrease in the dictator's internal sanctions against violence would be expected to increase the amount of time allocated to violence by the dictator. Further, if the level of fines and other monetary costs imposed by external agencies (e.g., the courts) as a result of the family violence do not vary with the level of violence, then the model indicates that an increase in such monetary sanctions will cause a reduction in the amount of time the dictator allocates to violence. If both the dictator and victim are risk neutral, an increase in the probability of external intervention will decrease the time allocated to violence. In addition, it is found that increases in the opportunities available to the victim outside the marriage will tend to improve the well-being of the victim in the marriage even if it has no effect on the time allocated to violence by the dictator. The model also provides insights for empirical work in family violence such as (1) suggestions of relevant independent variables, (2) the specification of a functional form for estimation, and (3) the specification of an error structure for the empirical model.     

Judgments of the Carter-Reagan Debate: The Eyes of the Beholders

In line with social-psychological theory and with findings fromprior analyses of the impact of televised campaign debates,the 1980 Carter-Reagan debate seems to have been interpretedlargely in light of preexisting candidate preferences. Knowledgeof whether a viewer intended to vote for Carter or Reagan and,for undecided voters, knowledge of position on the liberal-conservativecontinuum, permit accurate predictions to be made of judgmentsof who won the debate.     

Dependent masking and system life data analysis: Bayesian inference for two-component systems

Data from field operations of a system is often used to estimate the reliability of components. Under ideal circumstances, this system field data contains the time to failure along with information on the exact component responsible for the system failure. However, in many cases, the exact component causing the failure of the system cannot be identified, and is considered to be masked. Previously developed models for estimation of component reliability from masked system life data have been based upon the assumption that masking occurs independently of the true cause of system failure. In this paper we develop a Bayesian methodology for estimating component reliabilities from masked system life data when the probability of masking is dependent upon the true cause of system failure. The Bayesian approach is illustrated for the case of a two-component system of exponentially distributed components.     

Policy simulations with a macroeconometric model of the Indian economy

In this paper, the results of various dynamic policy simulations are analyzed within the context of a macroeconometric model of the Indian economy. The model contains 35 equations and offers a consistent framework for policy analysis. It is considerably expanded on the side of the fiscal sector and usefully incorporates the interdependence between monetary and fiscal sectors and gives due attention to supply side considerations. Magnitudes of effects of sustained policy changes are analyzed for the period 1964–1965 to 1974–1975. Impact multipliers and elasticities are also analyzed. Government expenditures variables and deficit financing are shown to have substantial impact on the system whereas changes in tax-rates, discount-rates, and liquidity ratios for commercial banks are shown to have only a marginal impact. The model is used for exploring the growth potential of the economy in a forecast period of five years under alternative assumptions regarding policy options.     

ASPECTS OF CORRELATION IN BIVARIATE POISSON DISTRIBUTIONS AND PROCESSES

This paper reviews some interesting but scattered results that are known about correlation in bivariate Poisson distributions and processes and presents some new results. A particular concern in both contexts is with results and examples relating to negative correlation.