Discrete Response Research Articles

An interpretable neural network-based non-proportional odds model for ordinal regression

This study proposes an interpretable neural network-based non-proportional odds model (N3POM) for ordinal regression. N3POM is different from conventional approaches to ordinal regression with non-proportional models in several ways: (1) N3POM is defined for both continuous and discrete responses, whereas standard methods typically treat the continuous variables as if they were discrete, (2) instead of estimating response-dependent finite-dimensional coefficients of linear models from discrete responses as is done in conventional approaches, we train a non-linear neural network to serve as a coefficient function. Thanks to the neural network, N3POM offers flexibility while preserving the interpretability of conventional ordinal regression. We establish a sufficient condition under which the predicted conditional cumulative probability locally satisfies the monotonicity constraint over a user-specified region in the covariate space. Additionally, we provide a monotonicity-preserving stochastic (MPS) algorithm for effectively training the neural network. We apply N3POM to several real-world datasets.

Penerapan Model Regresi Hurdle Binomial Negatif Menggunakan Algoritma Broyden-Fletcher-Goldfarb-Shanno pada Data Jumlah Kematian Bayi di Kota Makassar Tahun 2017

Poisson regression is a nonlinear regression method used to analyse the relationship between discrete response variables. Equidispersion is the assumption that must be met in the Poisson regression. Furthermore, there are cases in which the equidispersion assumption is invalidated when using the Poisson regression model to analyze data. One such case is overdispersion, which occurs when there is an excess of zero. As a result, the Negative Hurdle Binomial (HBN) regression is implemented to solve the overdispersion issue. Maximum Likelihood Estimation (MLE) with the Broyden-Fletcher-Goldfarb-Shanno (BFGS) algorithm was applied in this study to perform parameter estimation. In addition, the HBN regression model was used to analyze the data on the number of infant mortality cases in Makassar in 2017 with the variables assumed to be significant with infant mortality. The percentage of infants who were exclusively breastfed was the variable that had a significant impact on the outcome of HBN regression on the data on the number of infant mortality that experienced overdispersion.

A Dynamic Analysis of Arrears and Income Poverty in Italy

AbstractWe explore the effects of income poverty on the probability of being in arrears in Italy. We account for endogeneity by using a bivariate discrete response model, which allows considering feedback effects from arrears to future poverty status and selection issues. We also offer an analysis by macro-region. We use longitudinal 2016–2019 EU-SILC data. Our results suggest that being in arrears is characterized by a significant trap effect, which, however, tends to decline over time. We also find the presence of feedback effects from arrears condition to future poverty status. Notably, current poverty status is negatively associated with current arrears status, but the correlation is positive in the medium term. This may suggest that income conditions need time to exert their effects on the probability of being in arrears. In the medium term, being in arrears increases the probability of being poor. We also note heterogeneity at the regional level.

Intrinsic dual emissive insulin capped Au/Ag nanoclusters as single ratiometric nanoprobe for reversible detection of pH and temperature and cell imaging

pH and temperature are two important characteristics in cells and the environment. These, not only in the well-done regulation of cell functions but also in diagnosis and treatment, have a key role. Protein-protected bimetallic nanoclusters are abundantly used in the building of biosensors. However, insulin-stabilized Au–Ag nanoclusters with dual intrinsic emission have not been investigated yet. In this work, Dual emissive insulin templated Au–Ag nanocluster (Ins(Au/Ag)NCs) were first synthesized in a simple and green one-put manner. The two emission wavelengths of, as-prepared NCs centered at 410 and 630 nm, excited in one excitation wavelength (330 nm). These two emission peaks were assigned to the di-Tyrosine cross-linked formation and bimetallic nanoclusters respectively. Further analysis displayed that each emission band of Ins(Au/Ag)NCs responded to one variable whilst another peak remained constant; For blue and red emission wavelengths, pH dependency and thermo-responsibility were observed respectively. As-prepared nanoprobe with the intrinsic dual emissive feature was used for ratiometric determination of these parameters, each with a discrete response from another. The linear range of 6.0–9.0 for pH and 1 to 71 °C for temperature was obtained, which comprises the physiological range of pH and temperature and afforded intracellular sensing and imaging capability. As-prepared NCs probe show excellent biocompatibility and cell membrane permeability, and so were successfully applied as direct ratiometric pH and temperature probes in HeLa and HFF cells. More interestingly, this dual emissive nanoprobe is capable of distinguishing cancer cells from normal ones.

Modelling of Poor Household in East Kalimantan Using Zero Inflated Ordered Probit (ZIOP) Approach

Discrete ordinal response variables often exhibit an “excess” of zeros, which can be attributed to two different data conditions. Conventional ordinal probit models are limited in their ability to explain these excess zeros. The Zero Inflated Ordered Probit (ZIOP) model, which combines binary probit and ordinal probit regression, can be used to address this limitation. This study reviews the empirical analysis of the factors that contribute to poor household levels in East Kalimantan. The goodness-of-fit of the ZIOP and ordinal probit models was assessed using the Vuong test.

Bayesian estimation for median discrete Weibull regression model

<abstract><p>The discrete Weibull model can be adapted to capture different levels of dispersion in the count data. This paper takes into account the direct relationship between explanatory variables and the median of discrete Weibull response variable. Additionally, it provides the Bayesian estimate of the discrete Weibull regression model using the random walk Metropolis algorithm. The prior distributions of the coefficient predictors were carried out based on the uniform non-informative, normal and Laplace distributions. The performance of the Bayes estimators was also compared with the maximum likelihood estimator in terms of the mean square error and the coverage probability through the Monte Carlo simulation study. Meanwhile, a real data set was analyzed to show how the proposed model and the methods work in practice.</p></abstract>

Performance Evaluation of Jerusalem Artichoke Digging Tool in Cohesive Soil Using Discrete Element Method

Highlights Response surface methodology is suitable for DEM input parameter optimization. Soil reaction forces reduced at velocity ratios greater than one (1.2-3.9). Vibration reduced soil reaction forces at the target depth of 350 mm by 70%. In general, soil reaction forces increase with speed but decrease with frequency. Abrasive wear predominantly occurred at the tool’s cutting section. Abstract. The discrete element method (DEM) and response surface methodology (RSM) were used to determine the input parameters and combination of operational factors required for optimizing the Jerusalem artichoke (Helianthus tuberosus L.) harvesting tool in cohesive soil. The DEM soil model consisted of particles with different radii in three shapes calibrated using angle of repose and cone penetration data. Compared with data from a soil bin subsoiler evaluation, the DEM model showed acceptable relative errors for draught force (6.7%), vertical force (4.5%), and furrow width (9.3%). The effects of operational factors, including forward speed, vibration frequency, and amplitude, on response variables such as draught and vertical forces, drawbar power, and abrasive wear were analyzed for three harvesting shovels (S-shape, step-shape, and fork-shape). The ratio of vibratory speed to forward speed (velocity ratio, Vr) was used to analyze the combined effect of the factors. The operational factors significantly affected all the response variables (p<0.05). At Vr > 1 (1.2-3.9), soil reaction forces and drawbar power were considerably reduced. The optimal parameters for minimizing the response variables were 2.5 km h-1 forward speed, 14.5 Hz frequency, 30 mm amplitude, and S-shape shovel at Vr = 3.9. The minimum draught force, vertical force, drawbar power, and Archard wear depth were 4.64 kN, 0.41 kN, 2.64 kW, and 2.36 mm, respectively, at an operating depth of 350 mm. Operating in vibratory mode reduced draught force by 54% with the full width of the implement. Future work should include Jerusalem artichoke tubers in the simulation and experimental validation. Keywords: Abrasive wear, Clay, Numerical optimization, Soil reaction forces, Velocity ratio, Vibration.

A single-step recursive algorithm of the convolution integral for computing non-viscous damping forces in dynamic analyses

The non-viscous damping force features the convolution integral of the velocity and the kernel function, causing the difficulty of the solutions of the system equation of motion due to the nonlocal time dependence. In accordance with the linear time-invariant system theory in signal processing, this paper presents a single-step recursive algorithm of convolution integral for efficiently computing non-viscous damping forces. The algorithm parameters are effectively identified from the discrete responses of the kernel function to the rectangular unit impulse through digital filter technologies. The proposed algorithm is compatible with the solution schemes based on existing time integration methods, which handle the nonlocal dependence using internal recursive support vectors while leaving the order of the equation of motion unchanged. The central difference method, the Newmark-ß method, and the Bathe method are employed to illustrate the use of the algorithm, proving to retain the stability characteristics of time integration methods. The single and multiple DOF examples show that the simulated results based on the compatible solution schemes compare well with those built on other existing methods, verifying high accuracy and efficiency. The proposed algorithm thus extends the application range of existing time integration methods from the viscously damped system to the non-viscously damped one, fully demonstrating its generalizability.

Dynamic fit index cutoffs for categorical factor analysis with Likert-type, ordinal, or binary responses.

Scale validation is vital to psychological research because it ensures that scores from measurement scales represent the intended construct. Fit indices are commonly used to provide quantitative evidence that a proposed factor structure is plausible. However, there is a mismatch between guidelines for evaluating fit of the factor models and the data that most researchers have. Namely, fit guidelines are based on the simulations that assume item responses are collected on a continuous scale whereas most researchers collect discrete responses such as with a Likert-type scale. In this article, we show that common guidelines derived from assuming continuous responses (e.g., root-mean-square error of approximation < 0.06, comparative fit index > 0.95) do not generalize to factor models applied to discrete responses. Specifically, discrete responses provide less information than continuous responses, so less information about misfit is passed to fit indices. Traditional guidelines, therefore, end up being too lenient and lose their ability to identify that a model may have a poor fit. We provide one possible solution by extending the recently developed dynamic fit index framework to accommodate discrete responses common in psychology. We conduct a simulation study to provide evidence that the proposed method consistently distinguishes between well-fitting and poorly fitting models. Results showed that our proposed cutoffs maintained at least 90% sensitivity to misspecification across studied conditions, whereas traditional cutoffs were highly inconsistent and frequently exhibited sensitivity below 50%. The proposed method is included in the dynamic R package and as a web-based Shiny application to make it easily accessible to psychologists. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

Probabilistic and rich individual working memories revealed by a betting game

When asked to remember a color, do people remember a point estimate (e.g., a particular shade of red), a point estimate plus an uncertainty estimate, or are memory representations rich probabilistic distributions over feature space? We asked participants to report the color of a circle held in working memory. Rather than collecting a single report per trial, we had participants place multiple bets to create trialwise uncertainty distributions. Bet dispersion correlated with performance, indicating that internal uncertainty guided bet placement. While the first bet was on average the most precisely placed, the later bets systematically shifted the distribution closer to the target, resulting in asymmetrical distributions about the first bet. This resulted in memory performance improvements when averaging across bets, and overall suggests that memory representations contain more information than can be conveyed by a single response. The later bets contained target information even when the first response would generally be classified as a guess or report of an incorrect item, suggesting that such failures are not all-or-none. This paradigm provides multiple pieces of evidence that memory representations are rich and probabilistic. Crucially, standard discrete response paradigms underestimate the amount of information in memory representations.

Abstract 251: Approach to Prognostication for Patients Enrolled in a Post-Arrest Clinical Trial

Introduction/Background: Neuroprognostication after cardiac arrest is complex and the way clinicians prognosticate is not well understood. Research Question/Hypothesis: How do clinicians describe their approach to post-arrest prognostication in general and in specific patients they have treated? Goals/Aims: Characterize one or more prevailing themes in clinical decision-making about neuroprognostication. Methods/Approach: We surveyed and performed semi-structured interviews with clinicians who had cared for a subject enrolled in the Influence of Cooling Duration on Efficacy in Cardiac Arrest Patients (ICECAP) trial (NCT04217551). We performed interviews after trial subject died or was discharged. Our interviews explored clinician’s approach to neuroprognostication. Two authors (AS and YY) independently performed thematic analysis using inductive coding, then discussed any discrepancies and reached a consensus. Early in coding, we identified “clinical gestalt” as a common theme, defined as prognosticating based on patients overall appearance or impression of their history in the absence of any definitive prognostic tests. Each clinician provided their demographic information and local standard practices. We presented emerging themes and discrete responses with descriptive statistics as appropriate. We did not have access to subjects’ clinical identifiers or treatment arm of the trial. Results/Data: We interviewed 30 clinicians, who practiced at 19 US hospitals. Clinicians were primarily attendings (n=20) and most practiced critical care (n=21). Many (21/30) clinicians reported using clinical gestalt to formulate their early impressions. Of these, 9/21 reported high certainty in their initial predictions. Clinical gestalt informed both prediction of good and poor outcomes. Clinicians described different ways of incorporating gestalt into prognostication, including arrest history (17/21), severity of initial shock or multi-system organ failure (13/21), past clinical experience (10/21), comorbidities (10/21), and patient physical appearance (6/21). Conclusion: Clinicians often use gestalt when formulating early impressions of patient’s prognosis, an approach rarely discussed in prognostication literature.

Relationship between FAMACHA© scores and zootechnical indicators of a sheep production system

In Brazil, lamb producers face challenges raising their animals because of high anthelmintic resistance and loss of productivity due to parasites. It is well known that parasitic infections can reduce the performance of sheep. However, until the publication of this work, no research was found that quantified the effects that anemia reflected by FAMACHA© scores can exert on other zootechnical indicators in a lamb production system. The objective of this study was to use the FAMACHA© scores to quantify the impacts of anemia in ewes at breeding and lambing on the productive and reproductive performance of a meat sheep flock. The variables evaluated were i) FAMACHA© score of ewes at breeding and lambing, ii) body condition score of ewes at breeding and at lambing, iii) average daily gain of lambs until weaning, iv) ewe’s age, v) birth weights per individual lamb and litter, vi) weaning weight, vii) ewe’s weight, viii) litter size and ix) pre-weaning survival. The treatments evaluated corresponded to the FAMACHA© score of the ewes during breeding and lambing. Quantitative responses were submitted to analyses of variance and compared by Duncan’s test. In contrast, qualitative or discrete responses were evaluated by the Kruskal &amp; Wallis test and compared to Dunn’s test. The Wilcoxon test was performed to compare the FAMACHA© scores of ewes at breeding and lambing. All statistical analyzes were performed using the R-Studio software version 4.2.0 at a 5% significance level. The FAMACHA© score of breeding ewes was related to the body condition score at breeding, ewe weight, pre-weaning survival, ewe age, litter size, and birth weight. Furthermore, the FAMACHA© score of ewes at lambing was related to the body condition score at lambing, ewe weight, pre-weaning survival, birth weight per lamb and litter, pre-weaning average daily gain, weaning weight, and age of the ewe. There was no significant difference between the test times of the FAMACHA© scores of the ewes, indicating that an ewe will present a similar score at both stages. It was concluded that ewes with FAMACHA© scores of 4 and 5 and their offspring showed the worst productive and reproductive performances. Conversely, the ewes with FAMACHA© 1 obtained the opposite result, demonstrating better technical performance.

Simulation and analysis of soil homogenization drills based on discrete element method and response surface methodology

Currently, rotary drilling is one of the main pieces of equipment used for in-situ remediation of contaminated soil. However, this equipment has problems such as uneven mixing and low utilization efficiency, which affect the efficiency of in-situ soil remediation. To improve the efficiency of in-situ soil remediation, this paper takes contaminated black soil as the research object, and the structural design of the new three-stage soil remediation auger is carried out based on SolidWorks. The mixing process of soil and heavy metal passivator under different motion and structural parameters was investigated by the discrete element method (DEM) and response surface methodology. The experimental design was based on rotational speed, homogenizing mixing time, crushing section pitch, and homogenizing section pitch as factors, and soil fragmentation ratio, the coefficient of dispersion, and torque as optimization indices. The kinematic and structural parameters of the three-stage auger drill bit were then optimized using the one-factor method, the orthogonal test, and the response surface methodology, respectively. The test method uses a one-way test to determine the central level value of the orthogonal test and a comprehensive balance method to determine the best combination of parameters for the orthogonal test, which is then used as the central value of the response surface test for parameter optimization. The optimal combinations of kinematic and structural parameters of the three-stage auger drill bit are determined and validated using response surface methodology. The optimum combination of parameters was found to be a speed of 129 rpm, a homogenizing mixing time of 24 s, a pitch of 165 mm in the crushing section, and a pitch of 132 mm in the homogenizing section. The error between the optimal value of the predicted model using the response surface method and the actual simulated value under the optimal parameters is 4.2%, 4.9%, and 5.3%, respectively. The optimized factor parameters provide a reference for the design of the structural and kinematic parameters of the in-situ homogenization equipment.

Adaptive stochastic isogeometric analysis for nonlinear bending of thin functionally graded shells with material uncertainties

This paper presents an adaptive stochastic isogeometric method to incorporate material uncertainties in the nonlinear bending analysis of thin functionally graded material (FGM) shells. The gradient index is modeled as a second-order random field to describe the spatial randomness of material properties. An adaptive, nested, and non-intrusive Chebyshev interpolation process based on Leja sequences is formulated to approximate the response surface concerning random inputs, in which the discrete responses are parallelly obtained and incrementally augmented to save computational effort. The response surface is taken as a surrogate model to promote the efficiency of the stochastic analysis. In particular, a new probabilistic post-processing technique is proposed to accelerate the evaluation of the probability characteristics of the response. Isogeometric analysis is applied to solve the discrete responses of FGM shells, eliminating geometric errors and simplifying the implementation. Through three examples with different computational costs, the effectiveness of the present method is demonstrated by comparing it with two forms of Monte Carlo simulation methods. The influence of the statistical characteristics of the gradient index field on the response surface analysis and the random response is investigated.

Uncovering Novel Biomarkers of Inflammation as Potential Screening Targets of Disease Risk in Healthcare Shift Workers: A Pilot Study.

Shift work, experienced by nearly 30% of the U.S. workforce, is hazardous to health and has become a pervasive labor practice in the healthcare sector worldwide. It increases the risk of stroke, diabetes, cancer, and cardiovascular disease. Nonetheless, specific screening targets for shift workers still need to be defined. In this study, we have begun uncovering these targets as specific low-grade systemic inflammation markers and functional endotoxin-elicited responses that may foreshadow disease risk in shift workers. One hundred four participants (normothermic and normotensive) were healthy, non-smoking, and drug- and medication-free volunteers recruited from Atlanta area hospitals and medical schools. We assessed the concentration of three proteins in plasma samples from day workers and shift workers (lipopolysaccharide-binding protein, IL-10, and TNF-α), and the relationship between these baseline biomarkers and their response to an ex-vivo endotoxin challenge. We show that shift work increases low-grade systemic inflammation and disrupts discrete endotoxin responses. As shift work exposure increases, the correlation between low-grade systemic inflammation markers and their endotoxin responses was disrupted; this effect was more robust for TNF-α than for IL-10. With increased shift work exposure, these events, alone or combined, represent potential systemic and functional signals that may be harnessed to develop screening tools to identify at-risk individuals.

Discrete gust response analysis of SARAS aircraft

The objective of this work is to analyze the aerodynamic response of SARAS transport aircraft to a gust condition. A rigid airframe is assumed. The increments in load factors due to vertical gust are calculated for gust gradient distances varying from 30 to 350ft as per Federal Aviation Requirements (FAR) specifications. The variations in the angles of attack due to gust and airplane motion are also studied. This tuned discrete gust analysis includes the effects of unsteady aerodynamics. The analysis is performed considering the aircraft plunge degree of motion (tail-off). The effect of damping due to the tail contribution is also studied.

On the local modeling of count data: multiscale geographically weighted Poisson regression

A recent addition to the suite of techniques for local statistical modeling is the implementation of the multiscale geographically weighted regression (MGWR), a multiscale extension to geographically weighted regression (GWR). Using a back-fitting algorithm, MGWR relaxes the restrictive assumption in GWR that all processes being modeled operate at the same spatial scale and allows the estimation of a unique indicator of scale, the bandwidth, for each process. However, the current MGWR framework is limited to use with continuous data making it unsuitable for modeling data that do not typically exhibit a Gaussian distribution. This study expands the application of the MGWR framework to scenarios involving discrete response outcomes (count data following a Poisson’s distribution). Use of this new MGWR Poisson regression (MGWPR) model is demonstrated with a simulated data set and then with COVID-19 case counts within New York City at the zip code level. The results from the simulated data underscore the superiority of the MGWPR model in effectively capturing spatial processes that influence count data patterns, particularly those operating across diverse spatial scales. For empirical data, the results reveal significant spatial variations in relationships between socio-ecological factors and COVID-19 cases – variations often missed by traditional ‘global’ models.

Quality assessment in a multi-level control system with discrete response functions

The article extends the authors' previous research to the case of discrete response functions. A process consisting of sequential operations is considered. For each operation, you can determine the currently available probability of its error-free (high-quality) execution. By the qualitative performance of the operation, we will understand the achievement of a given level of quality in quantitative proportion to the overall level. The purpose of the article is to build a quality management strategy in which a set of probabilities of quality levels not less than a certain threshold value is achieved, with minimal costs for carrying out appropriate measures. The corresponding optimization problem is presented and a heuristic multistep algorithm for its solution is given.

층화 조건부 이산양적 확률화응답모형

When the population is composed of several strata and the information we want to estimate is a sensitive discrete quantitative character we propose a stratified conditional discrete quantitative randomized response model using less than sensitive character. First, we apply stratified sampling to the model of Carr, Marascuilo.(1982) which used less than sensitive character and look the results of obtaining sensitive qualitative information. Then, we applly stratified sampling to the model of Lee, Hong, Son.(2009) which used less than sensitive character as a condition to obtain information for a sensitive discrete quantitative character and establish a theoretical system for obtaining quantitative sensitive information by stratum. We dealt with the proportional and optimal allocations are examined as a method of allocating samples to each stratum. We also propose a stratified Liu-Chow model by applying stratified sampling and compare the efficiency between the suggested two models. Finally, in case of numerical comparison between two models, we find that the values of ph and th are more increasing the stratified conditional discrete quantitative randomized response model is the more efficiency than the stratified Liu-Chow model.

How to develop, test, and extend multinomial processing tree models: A tutorial.

Many psychological theories assume that observable responses are determined by multiple latent processes. Multinomial processing tree (MPT) models are a class of cognitive models for discrete responses that allow researchers to disentangle and measure such processes. Before applying MPT models to specific psychological theories, it is necessary to tailor a model to specific experimental designs. In this tutorial, we explain how to develop, fit, and test MPT models using the classical pair-clustering model as a running example. The first part covers the required data structures, model equations, identifiability, model validation, maximum-likelihood estimation, hypothesis tests, and power analyses using the software multiTree. The second part introduces hierarchical MPT modeling which allows researchers to account for individual differences and to estimate the correlations of latent processes among each other and with additional covariates using the TreeBUGS package in R. All examples including data and annotated analysis scripts are provided at the Open Science Framework (https://osf.io/24pbm/). (PsycInfo Database Record (c) 2023 APA, all rights reserved).