A GEE-type approach to untangle structural and random zeros in predictors.
Ontology highlight
ABSTRACT:
SUBMITTER: Ye P
PROVIDER: S-EPMC6535372 | biostudies-literature | 2019 Dec
REPOSITORIES: biostudies-literature
Publications
A GEE-type approach to untangle structural and random zeros in predictors.
Ye Peng P Tang Wan W He Jiang J He Hua H
Statistical methods in medical research 20181126 12
Count outcomes with excessive zeros are common in behavioral and social studies, and zero-inflated count models such as zero-inflated Poisson (ZIP) and zero-inflated Negative Binomial (ZINB) can be applied when such zero-inflated count data are used as response variable. However, when the zero-inflated count data are used as predictors, ignoring the difference of structural and random zeros can result in biased estimates. In this paper, a generalized estimating equation (GEE)-type mixture model ...[more]
Similar Datasets
Project description:Count data with structural zeros are common in public health applications. There are considerable researches focusing on zero-inflated models such as zero-inflated Poisson (ZIP) and zero-inflated Negative Binomial (ZINB) models for such zero-inflated count data when used as response variable. However, when such variables are used as predictors, the difference between structural and random zeros is often ignored and may result in biased estimates. One remedy is to include an indicator of the structural zero in the model as a predictor if observed. However, structural zeros are often not observed in practice, in which case no statistical method is available to address the bias issue. This paper is aimed to fill this methodological gap by developing parametric methods to model zero-inflated count data when used as predictors based on the maximum likelihood approach. The response variable can be any type of data including continuous, binary, count or even zero-inflated count responses. Simulation studies are performed to assess the numerical performance of this new approach when sample size is small to moderate. A real data example is also used to demonstrate the application of this method.
| S-EPMC6426322 | biostudies-literature
Project description:This paper presents a peridynamics-based micromechanical analysis framework that can efficiently handle material failure for random heterogeneous structural materials. In contrast to conventional continuum-based approaches, this method can handle discontinuities such as fracture without requiring supplemental mathematical relations. The framework presented here generates representative unit cells based on microstructural information on the material and assigns distinct material behavior to the constituent phases in the random heterogenous microstructures. The framework incorporates spontaneous failure initiation/propagation based on the critical stretch criterion in peridynamics and predicts effective constitutive response of the material. The current framework is applied to a metallic particulate-reinforced cementitious composite. The simulated mechanical responses show excellent match with experimental observations signifying efficacy of the peridynamics-based micromechanical framework for heterogenous composites. Thus, the multiscale peridynamics-based framework can efficiently facilitate microstructure guided material design for a large class of inclusion-modified random heterogenous materials.
| S-EPMC7143128 | biostudies-literature
Project description:Continuous outcome data with a proportion of observations equal to zero (often referred to as semicontinuous data) arise frequently in biomedical studies. Typical approaches involve two-part models, with one part a logistic model for the probability of observing a zero and some parametric continuous distribution for modeling the positive part of the data. We propose a semiparametric model based on a biological system with competing damage manifestation and resistance processes. This allows us to derive a closed-form profile likelihood based on the retro-hazard function, leading to a flexible procedure for modeling continuous data with a point mass at zero. A simulation study is presented to examine the properties of the method in finite samples. We apply the method to a data set consisting of pulmonary capillary hemorrhage area in lab rats subjected to diagnostic ultrasound. Copyright © 2017 John Wiley & Sons, Ltd.
| S-EPMC5530377 | biostudies-literature
Project description:One reason for the widespread success of random forests (RFs) is their ability to analyze most datasets without preprocessing. For example, in contrast to many other statistical methods and machine learning approaches, no recoding such as dummy coding is required to handle ordinal and nominal predictors. The standard approach for nominal predictors is to consider all 2 k - 1 - 1 2-partitions of the k predictor categories. However, this exponential relationship produces a large number of potential splits to be evaluated, increasing computational complexity and restricting the possible number of categories in most implementations. For binary classification and regression, it was shown that ordering the predictor categories in each split leads to exactly the same splits as the standard approach. This reduces computational complexity because only k - 1 splits have to be considered for a nominal predictor with k categories. For multiclass classification and survival prediction no ordering method producing equivalent splits exists. We therefore propose to use a heuristic which orders the categories according to the first principal component of the weighted covariance matrix in multiclass classification and by log-rank scores in survival prediction. This ordering of categories can be done either in every split or a priori, that is, just once before growing the forest. With this approach, the nominal predictor can be treated as ordinal in the entire RF procedure, speeding up the computation and avoiding category limits. We compare the proposed methods with the standard approach, dummy coding and simply ignoring the nominal nature of the predictors in several simulation settings and on real data in terms of prediction performance and computational efficiency. We show that ordering the categories a priori is at least as good as the standard approach of considering all 2-partitions in all datasets considered, while being computationally faster. We recommend to use this approach as the default in RFs.
| S-EPMC6368971 | biostudies-literature
Project description:Single cell Hi-C techniques enable one to study cell to cell variability in chromatin interactions. However, single cell Hi-C (scHi-C) data suffer severely from sparsity, that is, the existence of excess zeros due to insufficient sequencing depth. Complicating the matter further is the fact that not all zeros are created equal: some are due to loci truly not interacting because of the underlying biological mechanism (structural zeros); others are indeed due to insufficient sequencing depth (sampling zeros or dropouts), especially for loci that interact infrequently. Differentiating between structural zeros and dropouts is important since correct inference would improve downstream analyses such as clustering and discovery of subtypes. Nevertheless, distinguishing between these two types of zeros has received little attention in the single cell Hi-C literature, where the issue of sparsity has been addressed mainly as a data quality improvement problem. To fill this gap, in this paper, we propose HiCImpute, a Bayesian hierarchical model that goes beyond data quality improvement by also identifying observed zeros that are in fact structural zeros. HiCImpute takes spatial dependencies of scHi-C 2D data structure into account while also borrowing information from similar single cells and bulk data, when such are available. Through an extensive set of analyses of synthetic and real data, we demonstrate the ability of HiCImpute for identifying structural zeros with high sensitivity, and for accurate imputation of dropout values. Downstream analyses using data improved from HiCImpute yielded much more accurate clustering of cell types compared to using observed data or data improved by several comparison methods. Most significantly, HiCImpute-improved data have led to the identification of subtypes within each of the excitatory neuronal cells of L4 and L5 in the prefrontal cortex.
| S-EPMC9232133 | biostudies-literature
Project description:Use of zero-inflated count data models is common in applications where the number of zero counts exceeds that predicted from a traditional count data model such as Poisson or negative binomial. When count data exhibiting inflated zero counts are correlated among subjects, a natural approach will be to fit a marginal model with the help of generalized estimating equations (GEE) that can incorporate subject-to-subject correlations. A GEE based zero-inflated negative binomial (ZINB) model is proposed to fit clustered counts with excessive zeros. However, the corresponding sandwich variance estimator appears to underestimate the true variance. The theoretical reasons for its failure are explained and a correction under additional modeling assumptions is offered. In addition, a clustered resampling (bootstrap) procedure is proposed to estimate the variance and it is shown that the bootstrap procedure captures the correct variance under no additional model assumptions. Utility of this marginal GEE based ZINB model over two other competing models has been assessed using a thorough simulation study. The resulting inference procedure is applied to study the association between the dental caries and fluoride exposures using a dataset extracted from the Iowa Fluoride Study. A number of risk factors of clinical significance are reliably identified using the proposed model.
| S-EPMC4303594 | biostudies-literature
Project description:Purpose of reviewAlthough type 2 diabetes (T2D) is one of the most important risk factors that leads to the development of de novo heart failure, there are limited data, particularly from a practical/qualitative standpoint, about predictors of heart failure in this population.Recent findingsSodium-glucose cotransporter-2 (SGLT-2) inhibitors have been shown to prevent the development of heart failure and the composite of heart failure and cardiovascular death in patients with T2D without known heart failure who have either established atherosclerotic vascular disease or multiple risk factors. The concept of primary prevention of heart failure has led many clinicians to inquire if there are specific risk/enrichment factors that may predict an increased risk of heart failure.SummaryIn this review, we identify some general and diabetes-specific risk factors that are associated with an increased risk of developing heart failure in people with T2D.
| S-EPMC6727884 | biostudies-literature
Project description:Background and objectivesDiabetes is the leading cause of kidney failure in the United States, but early structural determinants of renal function loss in type 2 diabetes are poorly defined. We examined the association between morphometrically determined renal structural variables and loss of renal function in 111 American Indians with type 2 diabetes who volunteered for a research kidney biopsy at the end of a 6-year clinical trial designed to test the renoprotective efficacy of losartan versus placebo. Participants were subsequently followed in an observational study, in which annual measurements of GFR (iothalamate) initiated during the clinical trial were continued.Design, setting, participants, & measurementsRenal function loss was defined as ≥40% loss of GFR from the research examination performed at the time of kidney biopsy. Associations with renal function loss were evaluated by Cox proportional hazards regression. Hazard ratios (HRs) were reported per 1-SD increment for each morphometric variable.ResultsOf 111 participants (82% women; baseline mean [±SD] age, 46 years old [±10]; diabetes duration, 16 years [±6]; hemoglobin A1c =9.4% [±2.2]; GFR=147 ml/min [±56]; median albumin-to-creatinine ratio, 41 mg/g [interquartile range, 13-158]), 51 (46%) developed renal function loss during a median follow-up of 6.6 years (interquartile range, 3.1-9.0). Fourteen had baseline GFR <90 ml/min, and three had baseline GFR <60 ml/min. Higher mesangial fractional volume (HR, 2.27; 95% confidence interval [95% CI], 1.58 to 3.26), percentage of global glomerular sclerosis (HR, 1.63; 95% CI, 1.21 to 2.21), nonpodocyte cell number per glomerulus (HR, 1.50; 95% CI, 1.10 to 2.05), glomerular basement membrane width (HR, 1.48; 95% CI, 1.05 to 2.08), mean glomerular volume (HR, 1.42; 95% CI, 1.02 to 1.96), and podocyte foot process width (HR, 1.28; 95% CI, 1.03 to 1.60); lower glomerular filtration surface density (HR, 0.62; 95% CI, 0.41 to 0.94); and fewer endothelial fenestrations (HR, 0.68; 95% CI, 0.48 to 0.95) were each associated with GFR decline after adjustment for baseline age, sex, duration of diabetes, hemoglobin A1c, GFR, and treatment assignment during the clinical trial.ConclusionsQuantitative measures of glomerular structure predict loss of renal function in type 2 diabetes.
| S-EPMC4741038 | biostudies-literature
Project description:We estimate generic statistical properties of a structural credit risk model by considering an ensemble of correlation matrices. This ensemble is set up by Random Matrix Theory. We demonstrate analytically that the presence of correlations severely limits the effect of diversification in a credit portfolio if the correlations are not identically zero. The existence of correlations alters the tails of the loss distribution considerably, even if their average is zero. Under the assumption of randomly fluctuating correlations, a lower bound for the estimation of the loss distribution is provided.
| S-EPMC4031172 | biostudies-literature
Project description:Amino acid-specific covalent labeling is well suited to probe protein structure and macromolecular interactions, especially for macromolecules and their complexes that are difficult to examine by alternative means, due to size, complexity, or instability. Here we present a detailed account of carbodiimide-based covalent labeling (with GEE tagging) applied to a glycosylated monoclonal antibody therapeutic, which represents an important class of biologic drugs. Characterization of such proteins and their antigen complexes is essential to development of new biologic-based medicines. In this study, the experiments were optimized to preserve the structural integrity of the protein, and experimental conditions were varied and replicated to establish the reproducibility and precision of the technique. Homology-based models were generated and used to compare the solvent accessibility of the labeled residues, which include D, E, and the C-terminus, against the experimental surface accessibility data in order to understand the accuracy of the approach in providing an unbiased assessment of structure. Data from the protein were also compared to reactivity measures of several model peptides to explain sequence or structure-based variations in reactivity. The results highlight several advantages of this approach. These include: the ease of use at the bench top, the linearity of the dose response plots at high levels of labeling (indicating that the label does not significantly perturb the structure of the protein), the high reproducibility of replicate experiments (<2 % variation in modification extent), the similar reactivity of the 3 target probe residues (as suggested by analysis of model peptides), and the overall positive and significant correlation of reactivity and solvent accessible surface area (the latter values predicted by the homology modeling). Attenuation of reactivity, in otherwise solvent accessible probes, is documented as arising from the effects of positive charge or bond formation between adjacent amine and carboxyl groups, the latter accompanied by observed water loss. The results are also compared with data from hydroxyl radical-mediated oxidative footprinting on the same protein, showing that complementary information is gained from the 2 approaches, although the number of target residues in carbodiimide/GEE labeling is fewer. Overall, this approach is an accurate and precise method for assessing protein structure of biologic drugs.
| S-EPMC4622050 | biostudies-literature