Impact Of Model Choice Research Articles

The impact of modeling choices on the assessment of Ni-Ti fatigue properties through surrogate specimens.

The implant of self-expandable Ni-Ti stents for the treatment of peripheral diseases has become an established medical practice. However, the reported failure in clinics highlights the open issue of the fatigue characterization of these devices. One of the most common approaches for calculating the Ni-Ti fatigue limit (commonly defined in terms of mean and alternate strain for a fixed number of cycles) consists of using surrogate specimens which replicate the strain distributions of the final device but in simplified geometries. The main drawback lies in the need for computational models to determine the local distribution and, hence, interpret the experimental results. This study aims at investigating the role of different choices in the model preparation, such as the mesh refinement and the element formulation, on the output of the fatigue analysis. The analyses show a strong dependency of the numerical results on modeling choices. The use of linear reduced elements enriched by a layer of membrane elements is successful to increase the accuracy of the results, especially when coarser meshes are used. Due to material nonlinearity and stent complex geometries, for the same loading conditions and element type, (i) different meshes result in different couples of mean and amplitude strains and (ii) for the same mesh, the position of the maximum mean strain is not coincident with the maximum amplitude, making difficult the selection of the limit values.

Impact of model choice in predicting urban forest storm damage when data is uncertain

Research that illuminates causes of urban forest storm damage is valuable for planning and management. However, logistical and safety concerns often delay post-storm surveys in urban areas; thus, surveys may include observations with unverified sources of damage. While this uncertainty is often ignored, it can make up a high proportion of the number of damaged trees. The goal of this research was to improve understanding of techniques for modeling storm damage in urban forests. Using urban forest storm damage inventories collected in Florida, post-Hurricane Irma (2017), we tested how different imputation methods, modeling procedures, and damage frequency levels could impact overall model results. We utilized machine learning algorithms Random Forest (RF) and k-Nearest Neighbors (KNN), and generalized linear models (GLM). We found that GLM and RF models gave overall unbiased predictions of damage across all methods and rarity levels, while KNN consistently under-predicted damage. Damage frequency influenced some measures of performance but did not impact variable significance. Imputation methods all identified consistent variables of most significance within each model procedure; however, there was variation among variables ranked moderately important. While both GLM and RF models identified plot tree basal area as highly significant damage predictors, they otherwise disagreed on individual variable importance. These findings suggest that while explanatory models for urban forest storm damage can be achieved by examining linear relationships, more complex relationships such as the ones identified by RF models can have equal explanatory power with different subsets of predictor variables.

Impact of Model Choice When Studying the Relationship Between Blood Pressure Variability and Risk of Stroke Recurrence.

[Figure: see text].

Impact of model choice when studying the relationship between blood pressure variability and risk of stroke

Long-term blood pressure variability (BPV), an increasingly recognized vascular risk factor, is challenging to define and to analyze. In order to take into account two major methodological issues - conditioning on the future and uncertainty around BPV value - we compared different models on the risk of stroke in a large dataset. We used data from the PROGRESS trial, a secondary stroke prevention trial which included 6105 subjects followed-up during four years. The median number of BP measurements was 12 and 727 strokes occurred. We compared a naive Cox model including BPV as a fixed covariate calculated on the entire follow-up, to other models aimed at avoiding conditioning on the future and taking into account uncertainty around BPV value. We found that BPV was associated with an increased risk of stroke when using a naive Cox Model (HR = 1.19, 95% CI: 1.10–1.30) but not when using other models dealing more appropriately with the issue of conditioning on the future. These results illustrate that the methods used to estimate the association between BPV and stroke may affect the estimates and that more appropriate models tend to reduce this association. Figure: http://extranet.insight-outside.fr/upload/compte1352/Base/inscriptions_projets/image/99861-image.pdf

The impact of modelling choices on modelling outcomes: a spatio-temporal study of the association between COVID-19 spread and environmental conditions in Catalonia (Spain).

The choices that researchers make while conducting a statistical analysis usually have a notable impact on the results. This fact has become evident in the ongoing research of the association between the environment and the evolution of the coronavirus disease 2019 (COVID-19) pandemic, in light of the hundreds of contradictory studies that have already been published on this issue in just a few months. In this paper, a COVID-19 dataset containing the number of daily cases registered in the regions of Catalonia (Spain) since the start of the pandemic to the end of August 2020 is analysed using statistical models of diverse levels of complexity. Specifically, the possible effect of several environmental variables (solar exposure, mean temperature, and wind speed) on the number of cases is assessed. Thus, the first objective of the paper is to show how the choice of a certain type of statistical model to conduct the analysis can have a severe impact on the associations that are inferred between the covariates and the response variable. Secondly, it is shown how the use of spatio-temporal models accounting for the nature of the data allows understanding the evolution of the pandemic in space and time. The results suggest that even though the models fitted to the data correctly capture the evolution of COVID-19 in space and time, determining whether there is an association between the spread of the pandemic and certain environmental conditions is complex, as it is severely affected by the choice of the model.

Sphenodontian phylogeny and the impact of model choice in Bayesian morphological clock estimates of divergence times and evolutionary rates

BackgroundThe vast majority of all life that ever existed on earth is now extinct and several aspects of their evolutionary history can only be assessed by using morphological data from the fossil record. Sphenodontian reptiles are a classic example, having an evolutionary history of at least 230 million years, but currently represented by a single living species (Sphenodon punctatus). Hence, it is imperative to improve the development and implementation of probabilistic models to estimate evolutionary trees from morphological data (e.g., morphological clocks), which has direct benefits to understanding relationships and evolutionary patterns for both fossil and living species. However, the impact of model choice on morphology-only datasets has been poorly explored.ResultsHere, we investigate the impact of a wide array of model choices on the inference of evolutionary trees and macroevolutionary parameters (divergence times and evolutionary rates) using a new data matrix on sphenodontian reptiles. Specifically, we tested different clock models, clock partitioning, taxon sampling strategies, sampling for ancestors, and variations on the fossilized birth-death (FBD) tree model parameters through time. We find a strong impact on divergence times and background evolutionary rates when applying widely utilized approaches, such as allowing for ancestors in the tree and the inappropriate assumption of diversification parameters being constant through time. We compare those results with previous studies on the impact of model choice to molecular data analysis and provide suggestions for improving the implementation of morphological clocks. Optimal model combinations find the radiation of most major lineages of sphenodontians to be in the Triassic and a gradual but continuous drop in morphological rates of evolution across distinct regions of the phenotype throughout the history of the group.ConclusionsWe provide a new hypothesis of sphenodontian classification, along with detailed macroevolutionary patterns in the evolutionary history of the group. Importantly, we provide suggestions to avoid overestimated divergence times and biased parameter estimates using morphological clocks. Partitioning relaxed clocks offers methodological limitations, but those can be at least partially circumvented to reveal a detailed assessment of rates of evolution across the phenotype and tests of evolutionary mosaicism.

Influence of Modeling Choices on Value of Information Analysis: An Empirical Analysis from a Real-World Experiment.

Value of information (VOI) analysis often requires modeling to characterize and propagate uncertainty. In collaboration with a cancer clinical trial group, we integrated a VOI approach to assessing trial proposals. This paper aims to explore the impact of modeling choices on VOI results and to share lessons learned from the experience. After selecting two proposals (A: phase III, breast cancer; B: phase II, pancreatic cancer) for in-depth evaluations, we categorized key modeling choices relevant to trial decision makers (characterizing uncertainty of efficacy, evidence thresholds to change clinical practice, and sample size) and modelers (cycle length, survival distribution, simulation runs, and other choices). Using a $150,000 per quality-adjusted life-year (QALY) threshold, we calculated the patient-level expected value of sample information (EVSI) for each proposal and examined whether each modeling choice led to relative change of more than 10% from the averaged base-case estimate. We separately analyzed the impact of the effective time horizon. The base-case EVSI was $118,300 for Proposal A and $22,200 for Proposal B per patient. Characterizing uncertainty of efficacy was the most important choice in both proposals (e.g. Proposal A: $118,300 using historical data vs. $348,300 using expert survey), followed by the sample size and the choice of survival distribution. The assumed effective time horizon also had a substantial impact on the population-level EVSI. Modeling choices can have a substantial impact on VOI. Therefore, it is important for groups working to incorporate VOI into research prioritization to adhere to best practices, be clear in their reporting and justification for modeling choices, and to work closely with the relevant decision makers, with particular attention to modeling choices.

Impact of modelling choices on setting the reference levels for the EU forest carbon sinks: how do different assumptions affect the country-specific forest reference levels?

BackgroundIn 2018, the European Union (EU) adopted Regulation 2018/841, which sets the accounting rules for the land use, land use change and forestry (LULUCF) sector for the period 2021–2030. This regulation is part of the EU’s commitments to comply with the Paris Agreement. According to the new regulation, emissions and removals for managed forest land are to be accounted against a projected forest reference level (FRL) that is estimated by each EU Member State based on the continuation of forest management practices of the reference period 2000–2009. The aim of this study is to assess how different modelling assumptions possible under the regulation may influence the FRL estimates. Applying the interlinked G4M and WoodCarbonMonitor modelling frameworks, we estimate potential FRLs for each individual EU Member State following a set of conceptual scenarios, each reflecting different modelling assumptions that are consistent with the regulation and the technical guidance document published by the European Commission.ResultsThe simulations of the conceptual scenarios show that differences in the underlying modelling assumptions may have a large impact on the projected FRL. Depending on the assumptions taken, the projected annual carbon sink on managed forest land in the EU varies from −319 MtCO2 to −397 MtCO2 during the first compliance period (2021–2025) and from −296 MtCO2 to −376 MtCO2 during the second compliance period (i.e. 2026–2030). These estimates can be compared with the 2017 national GHG inventories which estimated that the forest carbon sink for managed forest land was −373 MtCO2 in 2015. On an aggregated EU level, the assumptions related to climate change and the allocation of forest management practices have the largest impacts on the FRL estimates. On the other hand, assumptions concerning the starting year of the projection, stratification of managed forest land, and timing of individual management activities are found to have relatively small impacts on the FRL estimates.ConclusionsWe provide a first assessment of the level of uncertainty associated with the different assumptions discussed in the technical guidance document and the LULUCF regulation, and the impact of these assumptions on the country-specific FRL. The results highlight the importance of transparent documentation by the EU Member States on how their FRL has been calculated, and on the underlying assumptions.

Bayesian Analysis of Multidimensional Item Response Theory Models: A Discussion and Illustration of Three Response Style Models

ABSTRACTInterest in Bayesian analysis of item response theory (IRT) models has grown tremendously due to the appeal of the paradigm among psychometricians, advantages of these methods when analyzing complex models, and availability of general-purpose software. Possible models include models which reflect multidimensionality due to designed test structure, construct-irrelevant variance, and mixed item formats. Using Markov Chain Monte Carlo methods, models can be estimated and evaluated for model-fit. In addition to discussing Bayesian analysis, analyses of three IRT models designed to account for extreme response style are illustrated: IRTree, multidimensional nominal response model (MNRM), and modified generalized partial credit model (MPCM). While results indicated there may be little impact of model choice on substantive trait estimates for the data set studied herein, model-fit results favored the MNRM and MPCM over the IRTree Model.

Modeling dynamics in diseased cardiac tissue: Impact of model choice.

Cardiac arrhythmias have been traditionally simulated using continuous models that assume tissue homogeneity and use a relatively large spatial discretization. However, it is believed that the tissue fibrosis and collagen deposition, which occur on a micron-level, are critical factors in arrhythmogenesis in diseased tissues. Consequently, it remains unclear how well continuous models, which use averaged electrical properties, are able to accurately capture complex conduction behaviors such as re-entry in fibrotic tissues. The objective of this study was to compare re-entrant behavior in discrete microstructural models of fibrosis and in two types of equivalent continuous models, a homogenous continuous model and a hybrid continuous model with distinct heterogeneities. In the discrete model, increasing levels of tissue fibrosis lead to a substantial increase in the re-entrant cycle length which is inadequately reflected in the homogenous continuous models. These cycle length increases appear to be primarily due to increases in the tip path length and to altered restitution behavior, and suggest that it is critical to consider the discrete effects of fibrosis on conduction when studying arrhythmogenesis in fibrotic myocardium. Hybrid models are able to accurately capture some aspects of re-entry and, if carefully tuned, may provide a framework for simulating conduction in diseased tissues with both accuracy and efficiency.

Impact of Model Choice on LR Assessment in Case of Rare Haplotype Match (Frequentist Approach)

AbstractThe likelihood ratio (LR) measures the relative weight of forensic data regarding two hypotheses. Several levels of uncertainty arise if frequentist methods are chosen for its assessment: the assumed population model only approximates the true one, and its parameters are estimated through a limited database. Moreover, it may be wise to discard part of data, especially that only indirectly related to the hypotheses. Different reductions define different LRs. Therefore, it is more sensible to talk about ‘a’ LR instead of ‘the’ LR, and the error involved in the estimation should be quantified. Two frequentist methods are proposed in the light of these points for the ‘rare type match problem’, that is, when a match between the perpetrator's and the suspect's DNA profile, never observed before in the database of reference, is to be evaluated.

Numerical simulation of tube-tooling cable-coating with polymer melts

This study investigates the numerical solution of viscous and viscoelastic flows for tube-tooling die-extrusion coating using a hybrid nite element/nite volume discretisation (fe/fv). Such a complex polymer melt extrusion-draw-coating flow displays a dynamic contact line, slip, die-swell and two separate free-surfaces, presenting an inner and outer conduit surface to the melt-coating. The practical interest lies in determining efficient windows for process control over variation in material properties, stressing levels generated and vacuum pressure levels imposed. The impact of shear-thinning is also considered. Extensive reference is made throughout to viscous inelastic counterpart solutions. Attention is paid to the influence and variation in relevant parameters of Weissenberg number (We), solvent-fraction (β) and second normal difference (N 2) (ξ parameter for EPTT). The impact of model choice and parameters upon field response is described in situ through, pressure-drops, rates of deformation and stress. Various numerical alternative strategies, their stability and convergence issues are also addressed. The numerical scheme solves the momentum-continuity-surface equations by a semi-implicit time-stepping Taylor-Galerkin/pressure-correction (TGPC) finite element (parent-cell) method, whilst invoking a sub-cell cell-vertex fluctuation distribution finite volume scheme for the constitutive stress equation. The hyperbolic aspects of the constitutive equation are addressed discretely through upwind Fluctuation Distribution techniques, whilst temporal and source terms are consistently accommodated through medium-dual-cell schemes. The dynamic solution of the moving boundary problem may be resolved by either separating the solution process for each free-surface section (decoupling), or coupling both sections and solving simultaneously. Each involves a surface height location method, with dependency on surface nodal velocities and surface element sections; two such schemes are investigated. Dedicated and localised shock-capturing techniques are introduced to handle solution singularities as disclosed by die-swell, slip and moving contact lines.

The impact of modelling choices in the predictive performance of richness maps derived from species‐distribution models: guidelines to build better diversity models

Summary The stacking of species‐distribution models (S‐SDMs) is receiving attention by conservation researchers because this approach is capable of simultaneously predicting species richness and composition. However, the steps required to build S‐SDMs implies at least two choices that influence its predictive performance which have not been extensively assessed: the selection of the modelling algorithm and the application of a threshold to transform the species‐distribution models into binary maps to be added together to build the final S‐SDM. Our goal was to provide guidelines concerning the best combinations of modelling algorithms and thresholds with which to build more accurate S‐SDMs. We generated 380 S‐SDMs of 1224 tree species in Mesoamerica by combining 19 distribution modelling methods with 20 different thresholds using presence‐only data from the Global Biodiversity Information Facility. We compared the predicted richness and composition with inventory data obtained from the BIOTREE‐NET forest plot database. We designed two indicators of predictive performance that were based on the diversity factors used to measure species turnover: a (shared species between the observed and predicted compositions), b and c (the exclusive species of the predicted and observed compositions respectively) and compared them with the Sorensen and Beta‐Simpson turnover measures. Our proposed indexes and the Sorensen index proved suitable as indicators of predictive performance for S‐SDMs, whereas the Beta‐Simpson turnover measure presented issues that would prevent its application to evaluate S‐SDMs. Some modelling methods – especially machine learning and ensemble model forecasting methods performed significantly better than others in minimizing the error in predicted richness and composition. Our results also points out that restrictive thresholds (with high omission errors) lead to more accurate S‐SDMs in terms of species richness and composition. Here, we demonstrate that particular combinations of modelling methods and thresholds provide results with higher predictive performance. These results provide clear modelling guidelines that will help S‐SDM modellers to select the appropriate combination of modelling methods and thresholds to build more accurate S‐SDMs, and therefore will have a positive impact on the quality of the diversity models used to assist conservation planning.

A Meta‐Analysis of Campaign Contributions' Impact on Roll Call Voting*

Objectives. Conventional wisdom about the link between campaign contributions and roll call votes is that contributions rarely matter because groups tend to give to like‐minded legislators. This meta‐analysis examines the conventional wisdom by analyzing published research on this topic.Methods. More than 30 studies are pooled to produce more than 350 individual tests of the contributions‐roll call link. Extending meta‐regression (Stanley and Jarrell, 1989), a logit meta‐analysis is conducted to summarize the literature and assess the importance of various modeling choices.Results. We find that some, but not all, model specifications have an impact on whether significant results are present. Models that control for friendly giving by including a measure of legislators' ideology and that include more than one contributions variable are less likely to produce significant results.Conclusions. After considering the impact of model choice on study results, we conclude that one‐third of roll call votes exhibit the impact of campaign contributions.

Lessons From a Comparison of Work Domain Models: Representational Choices and Their Implications

As methods in cognitive work analysis become more widely applied, questions regarding the impact of modeling choices and similarities in modeling efforts across projects and domains are increasingly relevant. However, no explicit comparison of models of similar systems has been reported. This paper compares independently developed work domain analysis (WDA) models of two command and control environments. Similarities in model content and the types of nodes included provide evidence that WDA techniques can capture fundamental elements regarding purposes and constraints. These points of agreement provide a common starting point for developing work domain representations of military command and control systems. The comparison also revealed differences between the models. Although differences in content reflected differences in scope of coverage and level of detail, other differences corresponded to more fundamental choices in modeling approach. These included the treatment of sensors, level of integration in the model, and representation of particular abstract constraints. Examination of these more fundamental differences pointed to important degrees of freedom in how to represent a WDA and clarified the implications of these modeling choices for guiding design. Actual or potential applications of this research include aiding analysts in making work domain modeling choices as well as producing work domain models of command and control environments.