Distribution Of Degeneration Research Articles

Inflated Mixture Models: Applications to Multimodality in Loss Given Default

In this paper, we propose an inflated mixture model to deal with multimodality in loss given default data. We propose a mixed of degenerate distributions, to handle zeros and ones excess, with a mixture of to-be-chosen bounded distributions for non-zeros and non-ones proportions. By applying the methodology in four retail portfolios of a large Brazilian commercial bank, we show that the inflated mixture of beta distributions plays better role minimizing model risk in fitting an inadequate model, in comparison with others considered competitive models. We explore the use of maximum likelihood estimation procedure. Monte Carlo simulations are carried out in order to check its finite sample performance.

A Tractable Monetary Model under General Preferences

This article studies an economy with both centralized and decentralized monetary exchanges under search frictions. A degenerate asset distribution is featured under a broad class of preferences including, for example, constant return to scale, constant elasticity of substitution, CARA and others from a range of macroeconomic literatures. Some novel applications impossible under quasi-linear preferences, for example endogenous growth, are illustrated under this class of preferences. This article finds that the welfare cost and growth loss of inflation can be much higher in these applications than previous estimates.

On performance of parametric and distribution-free models for zero-inflated and over-dispersed count responses.

Zero-inflated Poisson (ZIP) and negative binomial (ZINB) models are widely used to model zero-inflated count responses. These models extend the Poisson and negative binomial (NB) to address excessive zeros in the count response. By adding a degenerate distribution centered at 0 and interpreting it as describing a non-risk group in the population, the ZIP (ZINB) models a two-component population mixture. As in applications of Poisson and NB, the key difference between ZIP and ZINB is the allowance for overdispersion by the ZINB in its NB component in modeling the count response for the at-risk group. Overdispersion arising in practice too often does not follow the NB, and applications of ZINB to such data yield invalid inference. If sources of overdispersion are known, other parametric models may be used to directly model the overdispersion. Such models too are subject to assumed distributions. Further, this approach may not be applicable if information about the sources of overdispersion is unavailable. In this paper, we propose a distribution-free alternative and compare its performance with these popular parametric models as well as a moment-based approach proposed by Yu et al. [Statistics in Medicine 2013; 32: 2390-2405]. Like the generalized estimating equations, the proposed approach requires no elaborate distribution assumptions. Compared with the approach of Yu et al., it is more robust to overdispersed zero-inflated responses. We illustrate our approach with both simulated and real study data.

The Monte Carlo First-Come-First-Served Heuristic for Network Revenue Management

We introduce the Monte-Carlo based heuristic with first-come-first-served approximation for future optimal strategy (MC-FCFS) in order to maximize profit in a network revenue management problem. Like the randomized linear programming (RLP) model, one purpose of the MC-FCFS heuristic is to have information about displacement costs, considering the full probability distribution of future demands instead of a simplified degenerate distribution as in the deterministic linear programming (DLP) model. However, this information is conveyed by applying the FCFS heuristic as a future strategy rather than using the optimal ex-post profits as in the RLP heuristic. We show that MC-FCFS performs approximately as well as the RLP heuristic at a much lower computational cost and much better than the DLP heuristic at maximizing profit in a multi-night hotel booking setting with or without planned upgrades.

Predicting the EQ-5D-3L Preference Index from the SF-12 Health Survey in a National US Sample: A Finite Mixture Approach.

Background. When data on preferences are not available, analysts rely on condition-specific or generic measures of health status like the SF-12 for predicting or mapping preferences. Such prediction is challenging because of the characteristics of preference data, which are bounded, have multiple modes, and have a large proportion of observations clustered at values of 1. Methods. We developed a finite mixture model for cross-sectional data that maps the SF-12 to the EQ-5D-3L preference index. Our model characterizes the observed EQ-5D-3L index as a mixture of 3 distributions: a degenerate distribution with mass at values indicating perfect health and 2 censored (Tobit) normal distributions. Using estimation and validation samples derived from the Medical Expenditure Panel Survey 2000 dataset, we compared the prediction performance of these mixture models to that of 2 previously proposed methods: ordinary least squares regression (OLS) and two-part models. Results. Finite mixture models in which predictions are based on classification outperform two-part models and OLS regression based on mean absolute error, with substantial improvement for samples with fewer respondents in good health. The potential for misclassification is reflected on larger root mean square errors. Moreover, mixture models underperform around the center of the observed distribution. Conclusions. Finite mixtures offer a flexible modeling approach that can take into account idiosyncratic characteristics of the distribution of preferences. The use of mixture models allows researchers to obtain estimates of health utilities when only summary scores from the SF-12 and a limited number of demographic characteristics are available. Mixture models are particularly useful when the target sample does not have a large proportion of individuals in good health.

Population model-based optimization

Model-based optimization methods are a class of stochastic search methods that iteratively find candidate solutions by generating samples from a parameterized probabilistic model on the solution space. In order to better capture the multi-modality of the objective functions than the traditional model-based methods which use only a single model, we propose a framework of using a population of models at every iteration with an adaptive mechanism to propagate the population over iterations. The adaptive mechanism is derived from estimating the optimal parameter of the probabilistic model in a Bayesian manner, and thus provides a proper way to determine the diversity in the population of the models. We provide theoretical justification on the convergence of this framework by showing that the posterior distribution of the parameter asymptotically converges to a degenerate distribution concentrating on the optimal parameter. Under this framework, we develop two practical algorithms by incorporating sequential Monte Carlo methods, and carry out numerical experiments to illustrate their performance.

Asymptotic Distributions of the Generalized Range, Midrange, Extremal Quotient, and Extremal Product, with a Comparison Study

Necessary and sufficient conditions for the weak convergence of the generalized range, midrange, extremal quotient, and extremal product are obtained. The classes of possible non degenerate limit distribution functions of these simple statistics are characterized. Comparison study between these statistics with some examples for the most important distribution functions are given.

The correlation between ossification of the nuchal ligament and pathological changes of the cervical spine in patients with cervical spondylosis.

A retrospective clinical data analysis. This study was conducted to investigate the correlation between ossification of the nuchal ligament (ONL) and pathological changes of the cervical spine in patients with cervical spondylosis. ONL can usually be found in cervical spondylosis. Thus, it is important to find the correlation between ONL and pathological changes of the cervical spine in patients with cervical spondylosis. The medical records of 100 patients with cervical spondylosis with the local type of ONL (ONL group) and 50 patients with cervical spondylosis only (control group) were reviewed. Data analysis included patients' sex, age, location of ONL, maximum cord compression level, osteophyte height ratio, and grade of cervical intervertebral disc degeneration. Radiological features were evaluated by lateral plain radiography and magnetic resonance imaging. In total, 69.0% of subjects in the ONL group had ONL located at the maximum cord compression level; there was no difference based on sex (P = 0.248). The value of the osteophyte height ratio was higher at the ONL level than at its superior and inferior adjacent segments (P < 0.001). The osteophyte height ratio was also significantly different at the C4-C5 (P < 0.001) and C5-C6 (P = 0.008) levels between the ONL group and the control group. There was a significant difference in distribution of intervertebral disc degeneration grading between the ONL level and superior adjacent segments (P = 0.028), as well as inferior adjacent segments (P = 0.049). The distribution of intervertebral disc degeneration grading at the C5-C6 level between patients whose location of ONL and maximum cord compression level were both at C5-C6 and patients whose maximum cord compression level was at C5-C6 in the control group was also significantly different (P = 0.035). The location of ONL commonly corresponds to the most stenotic level of the spinal canal. The location of ONL also correlates with the level of osteophyte formation and intervertebral disc degeneration, indicating that ONL has correspondence to instability-related cervical pathological changes in cervical spondylosis.

Estimation of Parameters of a Mixed Failure Time Distribution Which is a Mixture of Degenerate (Degenerate at Zero) and I. G. Distribution

The problem of estimation of parameters of a mixture of degenerate (at zero) and exponential distribution is considered by Jayade and Prasad (1990). The sampling scheme proposed in it is extended in this paper to a mixture of degenerate and Inverse Gaussian distribution. The Inverse Gaussian distribution is very relevant for studying reliability and life-testing problems. The inverse Gaussian being the first passage time distribution for Wiener process makes it particularly appropriate for failure or reaction time data analysis.

Modulation of the Electronic Properties of Ultrathin Black Phosphorus by Strain and Electrical Field

The structural and electronic properties of the bulk and ultrathin black phosphorus and the effects of in-plane strain and out-of-plane electrical field on the electronic structure of phosphorene are investigated using first-principles methods. The computed results show that the bulk and few-layer black phosphorus from monolayer to six-layer demonstrates inherent direct bandgap features ranging from 0.5 to 1.6 eV. Interestingly, the band structures of the bulk and few-layer black phosphorus from X point via A point to Y point present degenerate distribution, which shows totally different partial charge dispersions. Moreover, strong anisotropy in regard to carrier effective mass has been observed along different directions. The response of phosphorene to in-plane strain is diverse. The bandgap monotonically decreases with increasing compressive strain, and semiconductor-to-metal transition occurs for phosphorene when the biaxial compressive reaches −9%. Tensile strain first enlarges the gap until the strai...

Inventory Sharing with Transshipment: Impacts of Demand Distribution Shapes and Setup Costs

We study a firm's optimal transshipment problem considering the impacts of setup costs for transshipment and demand distribution shapes. We assume that the demand follows a three‐point distribution, which changes from a degenerate distribution, to a unimodal distribution, and to a bimodal distribution as the demand shape parameter increases. We find that as the demand shape parameter increases, the optimal transshipment strategy changes from no transshipment to transshipment, and finally to no transshipment. The firm would use two‐way transshipment when the shape parameter is relatively small, while it would use one‐way transshipment when the shape parameter is relatively large. When the optimal strategy is one‐way transshipment, the transshipment direction depends on the contribution margin as well as the demand shape, when the difference between the two demand uncertainties is small. Our study of a dual‐channel retail system shows that the additional benefit of two‐way transshipment is negligible when there are many retail stores.

Central white matter degeneration in bulbar- and limb-onset amyotrophic lateral sclerosis.

Previous studies using diffusion tensor imaging (DTI) have examined for differences between bulbar- and limb-onset amyotrophic lateral sclerosis (ALS). Findings between studies have been markedly inconsistent, though possibly as a consequence of poor matching for confounding variables. To address this problem, this study contrasted the DTI profiles of limb-onset (ALS-L) and bulbar-onset (ALS-B) in groups that were tightly matched for the potential confounding effects of power, age, cognitive impairment and motor dysfunction. 14 ALS-L and 14 ALS-B patients were selected from a large prospective study so as to be matched on clinical and demographic features. All subjects, including 29 controls, underwent neuropsychological and neurological assessment. Tract-based spatial statistics and region of interest techniques were used to analyse fractional anisotropy (FA), mean diffusivity (MD), radial diffusivity (RD) and axial diffusivity (λ₁). Extensive bilateral FA and RD changes along the corticospinal tract were found in ALS-B compared to controls, p (corrected) <0.05; a similar distribution was seen for ALS-L at a less stringent statistical threshold. ROI analyses also showed more significant changes in ALS-B than ALS-L when each was compared to controls; for FA, MD and RD the changes reached statistical significance in the direct contrast between the two patient groups. With careful matching for confounding factors, the results suggest that ALS-B is associated with greater central white matter degeneration than ALS-L, possibly contributing to the known worse prognosis of ALS-B. The study, however, found no evidence that the spatial distribution of white matter degeneration differs between these groups.

Functional linear models for zero-inflated count data with application to modeling hospitalizations in patients on dialysis.

We propose functional linear models for zero-inflated count data with a focus on the functional hurdle and functional zero-inflated Poisson (ZIP) models. Although the hurdle model assumes the counts come from a mixture of a degenerate distribution at zero and a zero-truncated Poisson distribution, the ZIP model considers a mixture of a degenerate distribution at zero and a standard Poisson distribution. We extend the generalized functional linear model framework with a functional predictor and multiple cross-sectional predictors to model counts generated by a mixture distribution. We propose an estimation procedure for functional hurdle and ZIP models, called penalized reconstruction, geared towards error-prone and sparsely observed longitudinal functional predictors. The approach relies on dimension reduction and pooling of information across subjects involving basis expansions and penalized maximum likelihood techniques. The developed functional hurdle model is applied to modeling hospitalizations within the first 2 years from initiation of dialysis, with a high percentage of zeros, in the Comprehensive Dialysis Study participants. Hospitalization counts are modeled as a function of sparse longitudinal measurements of serum albumin concentrations, patient demographics, and comorbidities. Simulation studies are used to study finite sample properties of the proposed method and include comparisons with an adaptation of standard principal components regression.

Temperature-Dependent Hysteresis Behavior Modeling of the Open Circuit Potential and Impedance of LiFePO4 Batteries

Lithium Iron Phosphate (LFP) batteries present many challenges for accurate behavioral modeling during charge and discharge processes, as the two-phase interactions between cathode particles create multiple stable, degenerate distributions of lithium across the electrode. These degenerate distributions can cause the battery’s equilibrium open circuit potential (VOC) and equivalent series resistance (ESR) to vary at a given temperature and state of charge (SOC) depending on recent history, such as if the battery was most recently charged or discharged, a phenomenon known as hysteresis. A behavioral two-capacitor model was developed that incorporates temperature-dependent hysteresis between charge and discharge (affecting both VOC and ESR) and mimics temperature-dependent and SOC-dependent kinetics and impedance. The model includes a coupled finite volume thermal transport model with heat generation calculated due to a combination of Ohmic heating and entropic heating. The model was validated against high C-rate cyclic charge and discharge, as well as intermittent discharge and charge profiles at temperatures ranging from zero to 50°C. It is seen that for temperatures near 0°C, new physics come into play during recharge and that VOC hysteresis between charge and discharge increases as temperature decreases.

A spontaneous case of spinal cord degeneration in an adult dairy goat

A 2-year–old Saanen female goat, which showed a 10-day history of neurological deficit symptoms encompassing posterior weakness, difficulty in rising, and eventually inability to rise, was euthanatized because of poor prognosis. Histopathology revealed the presence of spinal cord degeneration characterized by Wallerian-type degeneration of myelinated nerve fibers in the white matter in association with myelin sheath degeneration, reactive astrocytosis, fibrillary astrogliosis, and capillary proliferation while sparing neurons in the gray matter horns. Lesions were bilateral and more marked in the lateral and ventral funiculi, particularly of the cervical and thoracic cord segments. Similar but far less severe degenerative lesions were present in the medulla oblongata, whereas other sites of the central nervous system, sciatic and femoral nerves, and skeletal muscles represented no significant changes. The nature and distribution of the spinal cord degeneration were distinguished from those reported previously in copper deficiency-induced syndrome or other neurodegenerative disorders in the goat. This condition was possibly a distinct variant of neurodegenerative disorder with specific involvement of the spinal cord in this species, although its etiology remains unknown.

Associated Factors of Intervertebral Disc Degeneration—Current Results from a Population-Based Cohort: The Wakayama Spine Study

Introduction Intervertebral disc degeneration (DD) is thought to be the first step of degenerative spinal changes. Furthermore, DD is considered to be one of the causes of several symptoms including low back pain (LBP). Although previous studies have investigated the association between DD and LBP as well as risk factors of DD, many of them are still controversial mainly because of the selection bias for the cohort. The purposes of this study were (1) to examine the association between the presence of DD and low back pain, and (2) to examine the risk factors related to DD (age, gender, body mass index [BMI], hyperlipidemia, diabetes mellitus [DM], peripheral artery disease [PAD], and smoking) using a large scale population-based cohort. Materials and Methods This was a cross-sectional study of the base-line data from the population-based cohort of Wakayama, western part of Japan, comprising 975 (men, 324; women, 651; mean age 66.4 years) of 1,011 participants (age, 21-97 years). The participants were subjected to the entire spine MRI and the lumbar intervertebral disc was evaluated according to the Pfirrmann classification system (1-3 = nondegeneration; 4, 5 = degeneration). Multiple logistic regression analysis was used to estimate the association between DD and low back pain after adjustment for age, gender, and BMI. Besides this, we analyzed the association between the presence of DD and the potential risk factors: age, gender, BMI, hyperlipidemia, diabetes mellitus (DM), peripheral artery disease (PAD), and smoking. Results The presence of DD was significantly associated with low back pain. The multiple logistic regression analysis indicated that older age, obesity, and hyperlipidemia were significant related factors of DD. Gender, DM, PAD, and smoking were not significantly related with DD in our cohort. Conclusion The current data provide the foundation for elucidating the causes and mechanisms of DD. However, there is a limitation concerning a temporal relationship between exposure and outcome due to the nature of cross-sectional study design. The second survey of the cohort, which is now ongoing, will provide important information in the near future. Disclosure of Interest None declared Reference Teraguchi M, Yoshimura N, Hashizume H, et al. Prevalence and distribution of intervertebral disc degeneration over the entire spine in a population-based cohort: the Wakayama Spine Study. Osteoarthritis Cartilage 2014;22(1):104–110

Ligation of mouse L4 and L5 spinal nerves produces robust allodynia without major motor function deficit

Spinal nerve L5/L6 ligation (SNL) in rats has become the standard for mechanistic studies of peripheral neuropathy and screening for novel analgesics. Conventional SNL in our hybrid mice resulted in a wide range of allodynia. Anatomical evaluation indicated that a variable number of lumbar vertebrae existed, resulting in L4/L5 or L5/L6 being ligated. Surprisingly, L4/L5 ligation did not result in ipsilateral hind limb paralysis and produced robust allodynia. Following a recent report that the mouse L4 neural segment is homologous with rat L5 we generated L4, L5 or both L4 and L5 (L4/L5) ligations in C57 mice after establishing a modified set of surgical landmarks. In contrast to rats, L4 ligation in these mice did not result in hind limb paralysis. Robust allodynia was observed in all three ligation groups. Nerve degeneration confirmed that L4 and L5, respectively, are primary contributors to the tibial and sural branches of the sciatic nerve in mice. A larger von Frey sensitive area reflected the wider distribution of Wallerian degeneration in the hindlimb of L4- compared to L5-ligated mice. Ligation of mouse L4 and L5 spinal nerves produces consistent, robust neuropathic pain behaviors and is suitable as a model for investigating mechanisms of neuropathic pain and for testing of novel analgesics. Gabapentin, used as a validation drug in neuropathic pain models and as a reference compound for novel analgesics, significantly reduced allodynia in the mice tested (L4/L5 ligations). Given the ease of surgery, robust allodynia, and larger von Frey sensitive area, we conclude that combined ligation of spinal nerves L4 and L5 optimizes the SNL model in mice.

Particle Filtering Framework for a Class of Randomized Optimization Algorithms

We reformulate a deterministic optimization problem as a filtering problem, where the goal is to compute the conditional distribution of the unobserved state given the observation history. We prove that in our formulation the conditional distribution converges asymptotically to a degenerate distribution concentrated on the global optimum. Hence, the goal of searching for the global optimum can be achieved by computing the conditional distribution. Since this computation is often analytically intractable, we approximate it by particle filtering, a class of sequential Monte Carlo methods for filtering, which has proven convergence in “tracking” the conditional distribution. The resultant algorithmic framework unifies some randomized optimization algorithms and provides new insights into their connection.

Prediction intervals for generalized-order statistics with random sample size

The problem of predicting future generalized-order statistics, by assuming the future sample size is a random variable, is discussed. A general expression for the coverage probability of the prediction intervals is derived. Since k-records and progressively type-II censored-order statistics are contained in the model of generalized-order statistics, the corresponding results for them can be deduced as special cases. When the future sample size has degenerate, binomial, Poisson and geometric distributions, numerical computations are given. The procedure for finding an optimal prediction interval is presented for each case. Finally, we apply our results to a real data set in life testing given in Lee and Wang [Statistical methods for survival data analysis. Hoboken, NJ: John Wiley and Sons; 2003, p. 58, Table 3.4] for illustrative the proposed procedure in this paper.

Diffusion tensor imaging comparison of progressive supranuclear palsy and corticobasal syndromes

BackgroundCorticobasal syndrome (CBS) and progressive supranuclear palsy syndrome (PSPS) are atypical parkinsonian syndromes that are both associated with white matter tract degeneration. However, little is known about how patterns of degeneration compare across these two syndromes. MethodsTwenty-seven subjects, nine with CBS and eighteen with probable or definite PSPS (9 pathologically confirmed) were prospectively recruited and underwent 3.0 T diffusion tensor imaging. A whole-brain voxel-based analysis was performed on fractional anisotropy (FA) and mean diffusivity (MD) images to compare both groups to each other and to 50 healthy controls. ResultsThe two syndromes showed overlapping regions of reduced FA and increased MD in the body of the corpus callosum, middle cingulum bundle, and premotor and prefrontal white matter, with reduced FA also observed in the superior cerebellar peduncles in both syndromes. However, CBS showed a more supratentorial and posterior pattern of degeneration with greater involvement of the splenium of the corpus callosum, premotor, motor and parietal lobes than PSPS. Findings in CBS were also highly asymmetric. Conversely, PSPS showed a more symmetric and infratentorial pattern of degeneration, with greater involvement of the superior cerebellar peduncles and midbrain than CBS. ConclusionsCBS and PSPS are both associated with striking white matter tract degeneration. Despite differences in the supratentorial and infratentorial distribution of degeneration, and in asymmetry, both tend to target a common structural network. Measurements of white matter tract diffusion could therefore be useful disease biomarkers in both of these syndromes.