In a 1992 Technometrics paper, Lambert (1992, 34, 1-14) described zero-inflated Poisson (ZIP) regression, a class of models for count data with excess zeros. In a ZIP model, a count response variable is assumed to be distributed as a mixture of a Poisson(lambda) distribution and a distribution with point mass of one at zero, with mixing probability p. Both p and lambda are allowed to depend on covariates through canonical link generalized linear models. In this paper, we adapt Lambert's methodology to an upper bounded count situation, thereby obtaining a zero-inflated binomial (ZIB) model. In addition, we add to the flexibility of these fixed effects models by incorporating random effects so that, e.g., the within-subject correlation and between-subject heterogeneity typical of repeated measures data can be accommodated. We motivate, develop, and illustrate the methods described here with an example from horticulture, where both upper bounded count (binomial-type) and unbounded count (Poisson-type) data with excess zeros were collected in a repeated measures designed experiment.

Excess weight in the form of fat mass does not provide additional benefits, and may potentially be negative, for adolescent bone.

Since species loss is predicted to be nonrandom, it is important to understand the manner in which those species that we anticipate losing interact with other species to affect ecosystem function. We tested whether litter species diversity, measured as richness and composition, affects breakdown dynamics in a detritus-based stream. Using full-factorial analyses of single- and mixed-species leaf packs (15 possible combinations of four dominant litter species; red maple [Acer rubrum], tulip poplar [Liriodendron tulipifera], chestnut oak [Quercus prinus], and rhododendron [Rhododendron maximum]), we tested for single-species presence/absence (additive) or species interaction (nonadditive) effects on leaf pack breakdown rates, changes in litter chemistry, and microbial and macroinvertebrate biomass. Overall, we found significant nonadditive effects of litter species diversity on leaf pack breakdown rates, which were explained both by richness and composition. Leaf packs containing higher litter species richness had faster breakdown rates, and antagonistic effects of litter species composition were observed when any two or three of the four litter species were mixed. Less-consistent results were obtained with respect to changes in litter chemistry and microbial and macroinvertebrate biomass. Our results suggest that loss of litter species diversity will decrease species interactions involved in regulating ecosystem function. To that end, loss of species such as eastern hemlock (Tsuga canadensis) accompanied by predicted changes in riparian tree species composition in the southeastern United States could have nonadditive effects on litter breakdown at the landscape scale.

Over the last decade or so, there has been increasing interest in 'zero inflated' (ZI) regression models to account for 'excess' zeros in data. Examples include ZI poisson (ZIP), ZI binomial (ZIB), ZI negative binomial and ZI tobit models. Recently, extensions of these models to the clustered data case have begun to appear. For example, Hall considered ZIP and ZIB models with cluster specific random effects. In this paper, we consider an alternative expectation maximization approach on the basis of marginal models and generalized estimating equation (GEE) methodology. In the usual EM algorithm for fitting ZI models, the M step is replaced by the solution of a GEE to take into account within cluster correlation. The details of this approach, including formulas for an asymptotic variance-covariance matrix of parameter estimates, are given for several of the most important ZI regression model classes. Alternatively, GEEs can be applied directly by computing the first two marginal moments of the observed response. We illustrate these two marginal modeling approaches with examples, and compare them via a small simulation study.

Context:Changes in serum vitamin D metabolites and calcium absorption with varying doses of oral vitamin D3 in healthy children are unknown.Objective:Our objective was to examine the dose-response effects of supplemental vitamin D3 on serum vitamin D metabolites and calcium absorption in children living at two U.S. latitudes.Design:Black and white children (n = 323) participated in a multisite (U.S. latitudes 34° N and 40° N), triple-masked trial. Children were randomized to receive oral vitamin D3 (0, 400, 1000, 2000, and 4000 IU/d) and were sampled over 12 weeks in winter. Serum 25-hydroxyvitamin D (25(OH)D) and 1,25-dihydroxyvitamin D (1,25(OH)2D) were measured using RIA and intact PTH (iPTH) by immunoradiometric assay. Fractional calcium absorption was determined from an oral stable isotope 44Ca (5 mg) in a 150-mg calcium meal. Nonlinear and linear regression models were fit for vitamin D metabolites, iPTH, and calcium absorption.Results:The mean baseline 25(OH)D value for the entire sample was 70.0 nmol/L. Increases in 25(OH)D depended on dose with 12-week changes ranging from −10 nmol/L for placebo to 76 nmol/L for 4000 IU. Larger 25(OH)D gains were observed for whites vs blacks at the highest dose (P < .01). Gains for 1,25(OH)2D were not significant (P = .07), and decreases in iPTH were not dose-dependent. There was no dose effect of vitamin D on fractional calcium absorption when adjusted for pill compliance, race, sex, or baseline 25(OH)D.Conclusion:Large increases in serum 25(OH)D with vitamin D3 supplementation did not increase calcium absorption in healthy children living at 2 different latitudes. Supplementation with 400 IU/d was sufficient to maintain wintertime 25(OH)D concentrations in healthy black, but not white, children.

No abstract

Loblolly pine (Pinus taeda L.) is the most important plantation species in the southeastern United States and specific gravity (SG) is its most important wood quality trait. Analysis of annual ring SG of breast height (1.37 m) increment cores from 3957 trees representing 147 plantations across the species natural range showed that ring SG increases with increasing age and varies significantly among physiographic regions. The South Atlantic and Gulf regions had the highest ring SGs, while the Hilly and Piedmont regions had the lowest. Based on ring SG, the juvenile period averages 4.3 years, followed by 8.6 years of transition wood, with mature wood produced by year 13. Whole-core mean SG was significantly higher in the South Atlantic (0.486) compared with the other regions (mean = 0.455), which were not statistically different from each other. Trees from the South Atlantic have significantly higher whole-core SG because they contain significantly more latewood (40.1%) compared with trees growing in other regions (33.8%). Maps indicate that stands in the South Atlantic and Gulf regions have the highest SG at a given age. Stands growing on the northern and western fringe of the natural range of loblolly pine have the lowest whole-core SG.Résumé : Le pin à encens (Pinus taeda L.) est la plus importante espèce plantée dans le sud-est des É tats-Unis d'Amérique et la densité est la plus importante caractéristique de la qualité du bois chez cette essence. L'analyse de la densité des cernes annuels dans des carottes prélevées à hauteur de poitrine (1,37 m) sur 3957 arbres provenant de 147 plantations établies partout dans l'aire naturelle de répartition de l'espèce montre que la densité des cernes annuels augmente avec l'âge et varie significativement entre les régions physiographiques. Les cernes annuels des régions de l'Atlantique Sud et du Golfe ont les densités les plus élevées tandis que ceux des régions de Hilly et du Piedmont ont les plus faibles. Sur la base de la densité des cernes annuels, la période juvénile dure en moyenne 4,3 ans, suivie d'une période de transition de 8,6 ans et du bois mature vers l'âge de 13 ans. La densité moyenne de l'ensemble de la carotte est significativement plus élevée dans la région de l'Atlantique Sud (0,486) que dans les autres régions (moyenne = 0,455) qui ne sont pas statistiquement différentes les unes des autres. Les carottes des arbres de la région de l'Atlantique Sud ont une densité significativement plus élevée parce qu'ils contiennent significativement plus de bois final (40,1 %) que les arbres qui croissent dans les autres régions (33,8 %). Pour un âge donné, les cartes indiquent que les peuplements des régions de l'Atlantique Sud et du Golfe ont la densité la plus élevée. Les carottes provenant des peuplements qui croissent aux limites nord et ouest de l'aire naturelle de répartition du pin à encens ont les plus faibles densités.[Traduit par la Rédaction]

Gymnasts' bone mineral characteristics are generally not known before starting their sport. Prepubertal females who enrolled in beginning artistic gymnastics (n = 65) had lower bone mineral than controls (n = 78). However, 2 years of gymnastics participation versus no participation led to a significantly greater accrual of forearm bone area and lumbar spine areal BMD.