Flexible Bivariate Count Data Regression Models

The article develops a semiparametric estimation method for the bivariate count data regression model. We develop a series expansion approach in which dependence between count variables is introduced by means of stochastically related unobserved heterogeneity components, and in which, unlike existing commonly used models, positive as well as negative correlations are allowed. Extensions that accommodate excess zeros, censored data, and multivariate generalizations are also given. Monte Carlo experiments and an empirical application to tobacco use confirms that the model performs well relative to existing bivariate models, in terms of various statistical criteria and in capturing the range of correlation among dependent variables. This article has supplementary materials online.     

The impact of female education on fertility: a natural experiment from Egypt

This paper presents new evidence on the impact of female education on fertility in Egypt using the change in the length of primary schooling as the source of exogenous variation in education. Beginning in 1988, the Egyptian government cut the number of primary school years from six to five, moving from a 12-year system of pre-university education to an 11-year system. This policy change affected all individuals born on or after October 1977. Using triennial pooled cross-section data from 1992 to 2014 and a nonparametric regression discontinuity approach, we compare education and fertility of women born just before and right after October 1977. Our analysis shows that female education significantly reduces the number of children born per woman. The reduction in fertility seems to result from delaying maternal age rather than changing women’s fertility preferences. We also provide evidence that female education in Egypt does not boost women’s labor force participation or affect their usages of contraceptive methods. Female education, however, does appear to increase women’s age at marriage which might explain the delay of maternal age.     

Youth Migration Decisions in Sub-Saharan Africa: Satellite-Based Empirical Evidence from Nigeria

This article examines the implications of urban growth on youth migration decisions in Nigeria. We use night light intensity data combined with Living Standards Measurement Study-Integrated Surveys on Agriculture data, as an indicator of urban growth and associated economic opportunities. Employing alternative econometric approaches that exploit the spatial and temporal differences in urban growth as proxied by night light intensity, we find that urban growth in potential migrant destinations encourages youth migration. We also find heterogeneous responses to urban growth among various groups of youth as well as varying responses to different types of migration. Broadly, women and those youth with more education are more likely to migrate, while those in households with livestock are less likely to migrate. Often, however, the effects are complex and varied. For example, land and physical asset ownership encourage temporary migration; but greater land ownership discourages permanent migration, while physical assets have insignificant effects. Our results from Nigeria show that policy makers concerned about rural–urban youth exodus should adopt a differentiated, in terms of targets, and multidimensional policy approach to reap the benefits of urbanization while avoiding its negative consequences.     

Africa's Rising Demand for Wheat: Trends,Drivers, and Policy Implications

This article analyses trends in wheat consumption and imports in sub‐Saharan Africa since 1980, and estimates the economic and demographic determinants of this rising demand for wheat. Results point to rising incomes, growing populations, and increasing women's labour‐force participation as key drivers. Urban wheat‐expenditure shares generally exceed rural ones and SSA's demand is met largely by imports and partly through domestic production on large‐scale farms. Rising demand may therefore entail few farm–non‐farm synergies and minimal prospects to spur broad‐based economic development. The article concludes by discussing policy options for African countries to meet their staple food needs while also promoting pro‐poor agricultural growth.     

Statistical Analysis of Uncertainties in Deterministic Computational Modeling – Application to Composite Process Resin Infusion Flow Model

Deterministic physics-based flow modeling provides an effective way to simulate and understand the resin flow infusion process in liquid composite molding processes and its variants. These are effective to provide optimal injection time and locations prior to gelation for given process parameters of resin viscosity and preform permeability. However, there could be significant variations in these two parameters during actual manufacturing. This paper presents simulation-based statistical analysis of uncertainties of these process parameters involved in the resin flow infusion. Two key process parameters, viscosity and permeability, and their statistical variations are examined individually and subsequently in combination for their impact on the associated injection time. Values from statistical probability distribution of the process parameters were employed to find the solution space for this engineering application through deterministic physics-based process flow modeling simulations. A bivariate confidence envelope was developed using the appropriate Cumulative Density Function for a 95% probability of successfully completing resin infusion prior to physical resin gelation time. A logistic regression model for the influence of resin viscosity and permeability on the binary response of successful resin infusion is presented and conforms well to the sensitivity analysis inferences.     

Tests for Detecting Overdispersion in the Positive Poisson Regression Model

This article derives score tests for extra-Poisson variation in the positive or truncated-at-zero Poisson regression model against truncated-at-zero negative binomial family alternatives. It also develops size-corrected tests of overdispersion that are expected to improve their small-sample properties. Further, small-sample performance of the tests is investigated by means of Monte Carlo experiments. As an illustration, the proposed tests are applied to a model of strikes in U.S. manufacturing. The proposed tests have an interpretation as conditional moment tests and require only the positive Poisson model to be estimated. It is shown that most of the tests for overdispersion in the regular Poisson model given in the econometric and statistical literature can be obtained as special cases of the tests developed in this article. Monte Carlo experiments indicate that the size correction, based on the asymptotic expansions of the score function, is effective in improving the accuracy of the size and power of the tests in small samples.