What do we gain from simplicity versus complexity in species distribution models?

Species distribution models (SDMs) are widely used to explain and predict species ranges and environmental niches. They are most commonly constructed by inferring species' occurrence-environment relationships using statistical and machine-learning methods. The variety of methods that can be used to construct SDMs (e.g. generalized linear/additive models, tree-based models, maximum entropy, etc.), and the variety of ways that such models can be implemented, permits substantial flexibility in SDM complexity. Building models with an appropriate amount of complexity for the study objectives is critical for robust inference. We characterize complexity as the shape of the inferred occurrence-environment relationships and the number of parameters used to describe them, and search for insights into whether additional complexity is informative or superfluous. By building 'under fit' models, having insufficient flexibility to describe observed occurrence-environment relationships, we risk misunderstanding the factors shaping species distributions. By building 'over fit' models, with excessive flexibility, we risk inadvertently ascribing pattern to noise or building opaque models. However, model selection can be challenging, especially when comparing models constructed under different modeling approaches. Here we argue for a more pragmatic approach: researchers should constrain the complexity of their models based on study objective, attributes of the data, and an understanding of how these interact with the underlying biological processes. We discuss guidelines for balancing under fitting with over fitting and consequently how complexity affects decisions made during model building. Although some generalities are possible, our discussion reflects differences in opinions that favor simpler versus more complex models. We conclude that combining insights from both simple and complex SDM building approaches best advances our knowledge of current and future species ranges.

Application of the european customer satisfaction index to postal services. Structural equation models versus partial least squars

Customer satisfaction and retention are key issues for organizations in today’s competitive market place. As such, much research and revenue has been invested in developing accurate ways of assessing consumer satisfaction at both the macro (national) and micro (organizational) level, facilitating comparisons in performance both within and between industries. Since the instigation of the national customer satisfaction indices (CSI), partial least squares (PLS) has been used to estimate the CSI models in preference to structural equation models (SEM) because they do not rely on strict assumptions about the data. However, this choice was based upon some misconceptions about the use of SEM’s and does not take into consideration more recent advances in SEM, including estimation methods that are robust to non-normality and missing data. In this paper, both SEM and PLS approaches were compared by evaluating perceptions of the Isle of Man Post Office Products and Customer service using a CSI format. The new robust SEM procedures were found to be advantageous over PLS. Product quality was found to be the only driver of customer satisfaction, while image and satisfaction were the only predictors of loyalty, thus arguing for the specificity of postal services

Eliciting consumers preferences using stated preference discrete choice models: Contingent ranking versus choice experiment

The aim of this paper is twofold: firstly, to carry out a theoreticalreview of the most recent stated preference techniques used foreliciting consumers preferences and, secondly, to compare the empiricalresults of two dierent stated preference discrete choice approaches.They dier in the measurement scale for the dependent variable and,therefore, in the estimation method, despite both using a multinomiallogit. One of the approaches uses a complete ranking of full-profiles(contingent ranking), that is, individuals must rank a set ofalternatives from the most to the least preferred, and the other usesa first-choice rule in which individuals must select the most preferredoption from a choice set (choice experiment). From the results werealize how important the measurement scale for the dependent variablebecomes and, to what extent, procedure invariance is satisfied.

Application of the european customer satisfaction index to postal services. Structural equation models versus partial least squars

Customer satisfaction and retention are key issues for organizations in today’s competitive market place. As such, much research and revenue has been invested in developing accurate ways of assessing consumer satisfaction at both the macro (national) and micro (organizational) level, facilitating comparisons in performance both within and between industries. Since the instigation of the national customer satisfaction indices (CSI), partial least squares (PLS) has been used to estimate the CSI models in preference to structural equation models (SEM) because they do not rely on strict assumptions about the data. However, this choice was based upon some misconceptions about the use of SEM’s and does not take into consideration more recent advances in SEM, including estimation methods that are robust to non-normality and missing data. In this paper, both SEM and PLS approaches were compared by evaluating perceptions of the Isle of Man Post Office Products and Customer service using a CSI format. The new robust SEM procedures were found to be advantageous over PLS. Product quality was found to be the only driver of customer satisfaction, while image and satisfaction were the only predictors of loyalty, thus arguing for the specificity of postal services

International evidence on the predictability of returns to securitized real estate assets: econometric models versus neural networks

The performance of various statistical models and commonly used financial indicators for forecasting securitised real estate returns are examined for five European countries: the UK, Belgium, the Netherlands, France and Italy. Within a VAR framework, it is demonstrated that the gilt-equity yield ratio is in most cases a better predictor of securitized returns than the term structure or the dividend yield. In particular, investors should consider in their real estate return models the predictability of the gilt-equity yield ratio in Belgium, the Netherlands and France, and the term structure of interest rates in France. Predictions obtained from the VAR and univariate time-series models are compared with the predictions of an artificial neural network model. It is found that, whilst no single model is universally superior across all series, accuracy measures and horizons considered, the neural network model is generally able to offer the most accurate predictions for 1-month horizons. For quarterly and half-yearly forecasts, the random walk with a drift is the most successful for the UK, Belgian and Dutch returns and the neural network for French and Italian returns. Although this study underscores market context and forecast horizon as parameters relevant to the choice of the forecast model, it strongly indicates that analysts should exploit the potential of neural networks and assess more fully their forecast performance against more traditional models.

The cross-currency hedging performance of implied versus statistical forecasting models

This article examines the ability of several models to generate optimal hedge ratios. Statistical models employed include univariate and multivariate generalized autoregressive conditionally heteroscedastic (GARCH) models, and exponentially weighted and simple moving averages. The variances of the hedged portfolios derived using these hedge ratios are compared with those based on market expectations implied by the prices of traded options. One-month and three-month hedging horizons are considered for four currency pairs. Overall, it has been found that an exponentially weighted moving-average model leads to lower portfolio variances than any of the GARCH-based, implied or time-invariant approaches.

Differential 3-year effects of first- versus second-generation antipsychotics on subjective well-being in schizophrenia using marginal structural models

This study examined the differential effects of first- (FGAs) versus second-generation antipsychotics (SGAs) on subjective well-being in patients with schizophrenia.

Clinical versus pre-clinical FE models for vertebral body strength predictions

The finite element analysis is an accepted method to predict vertebral body compressive strength. This study compares measurements obtained from in vitro tests with the ones from two different simulation models: clinical quantitative computer tomography (QCT) based homogenized finite element (hFE) models and pre-clinical high-resolution peripheral QCT-based (HR-pQCT) hFE models. About 37 vertebral body sections were prepared by removing end-plates and posterior elements, scanned with QCT (390/450μm voxel size) as well as HR-pQCT (82μm voxel size), and tested in compression up to failure. Non-linear viscous damage hFE models were created from QCT/HT-pQCT images and compared to experimental results based on stiffness and ultimate load. As expected, the predictability of QCT/HR-pQCT-based hFE models for both apparent stiffness (r2=0.685/0.801r2=0.685/0.801) and strength (r2=0.774/0.924r2=0.774/0.924) increased if a better image resolution was used. An analysis of the damage distribution showed similar damage locations for all cases. In conclusion, HR-pQCT-based hFE models increased the predictability considerably and do not need any tuning of input parameters. In contrast, QCT-based hFE models usually need some tuning but are clinically the only possible choice at the moment.

Simulation tools for statistical model comparison: an application to unobserved componenet models versus dynamic regression models

Aplicación de simulación de Monte Carlo y técnicas de Análisis de la Varianza (ANOVA) a la comparación de modelos estocásticos dinámicos para accidentes de tráfico.

Effects of potential models on the adsorption of ethane and ethylene on graphitized thermal carbon black. Study of two-dimensional critical temperature and isosteric heat versus loading

Adsorption of ethylene and ethane on graphitized thermal carbon black and in slit pores whose walls are composed of graphene layers is studied in detail to investigate the packing efficiency, the two-dimensional critical temperature, and the variation of the isosteric heat of adsorption with loading and temperature. Here we used a Monte Carlo simulation method with a grand canonical Monte Carlo ensemble. A number of two-center Lennard-Jones (LJ) potential models are investigated to study the impact of the choice of potential models in the description of adsorption behavior. We chose two 2C-LJ potential models in our investigation of the (i) UA-TraPPE-LJ model of Martin and Siepmann (J. Phys. Chem. B 1998,102, 25692577) for ethane and Wick et al. (J. Phys. Chem. B 2000,104, 8008-8016) for ethylene and (ii) AUA4-LJ model of Ungerer et al. (J. Chem. Phys. 2000,112, 5499-5510) for ethane and Bourasseau et al. (J. Chem. Phys. 2003, 118, 3020-3034) for ethylene. These models are used to study the adsorption of ethane and ethylene on graphitized thermal carbon black. It is found that the solid-fluid binary interaction parameter is a function of adsorbate and temperature, and the adsorption isotherms and heat of adsorption are well described by both the UA-TraPPE and AUA models, although the UA-TraPPE model performs slightly better. However, the local distributions predicted by these two models are slightly different. These two models are used to explore the two-dimensional condensation for the graphitized thermal carbon black, and these values are 110 K for ethylene and 120 K for ethane.

Adsorption of carbon tetrachloride on graphitized thermal carbon black and in slit graphitic pores: Five-site versus one-site potential models

The performance of intermolecular potential models on the adsorption of carbon tetrachloride on graphitized thermal carbon black at various temperatures is investigated. This is made possible with the extensive experimental data of Machin and Ross(1), Avgul et al.,(2) and Pierce(3) that cover a wide range of temperatures. The description of all experimental data is only possible with the allowance for the surface mediation. If this were ignored, the grand canonical Monte Carlo (GCMC) simulation results would predict a two-dimensional (2D) transition even at high temperatures, while experimental data shows gradual change in adsorption density with pressure. In general, we find that the intermolecular interaction has to be reduced by 4% whenever particles are within the first layer close to the surface. We also find that this degree of surface mediation is independent of temperature. To understand the packing of carbon tetrachloride in slit pores, we compared the performance of the potential models that model carbon tetrachloride as either five interaction sites or one site. It was found that the five-site model performs better and describes the imperfect packing in small pores better. This is so because most of the strength of fluid-fluid interaction between two carbon tetrachloride molecules comes from the interactions among chlorine atoms. Methane, although having tetrahedral shape as carbon tetrachloride, can be effectively modeled as a pseudospherical particle because most of the interactions come from carbon-carbon interaction and hydrogen negligibly contributes to this.

Hypoxic/ischemic models in newborn piglet: Comparison of constant FiO(2) versus variable FiO(2) delivery

A comparison of a constant (continuous delivery of 4% FiO(2)) and a variable (initial 5% FiO(2) with adjustments to induce low amplitude EEG (LAEEG) and hypotension) hypoxic/ischemic insult was performed to determine which insult was more effective in producing a consistent degree of survivable neuropathological damage in a newborn piglet model of perinatal asphyxia. We also examined which physiological responses contributed to this outcome. Thirty-nine 1-day-old piglets were subjected to either a constant hypoxic/ischemic insult of 30- to 37-min duration or a variable hypoxic/ischemic insult of 30-min low peak amplitude EEG (LAEEG < 5 mu V) including 10 min of low mean arterial blood pressure (MABP < 70% of baseline). Control animals (n = 6) received 21% FiO(2) for the duration of the experiment. At 72 h, the piglets were euthanased, their brains removed and fixed in 4% paraformaldehyde and assessed for hypoxic/ischemic injury by histological analysis. Based on neuropathology scores, piglets were grouped as undamaged or damaged; piglets that did not survive to 72 h were grouped separately as dead. The variable insult resulted in a greater number of piglets with neuropathological damage (undamaged = 12.5%, damaged = 68.75%, dead = 18.75%) while the constant insult resulted in a large proportion of undamaged piglets (undamaged = 50%, damaged = 22.2%, dead = 27.8%). A hypoxic insult varied to maintain peak amplitude EEG < 5 mu V results in a greater number of survivors with a consistent degree of neuropathological damage than a constant hypoxic insult. Physiological variables MABP, LAEEG, pH and arterial base excess were found to be significantly associated with neuropathological outcome. (c) 2006 Elsevier B.V. All rights reserved.

Three-factor models versus time series models:quantifying time-dependencies of interactions between stimuli in cell biology and psychobiology for short longitudinal data

Signal integration determines cell fate on the cellular level, affects cognitive processes and affective responses on the behavioural level, and is likely to be involved in psychoneurobiological processes underlying mood disorders. Interactions between stimuli may subjected to time effects. Time-dependencies of interactions between stimuli typically lead to complex cell responses and complex responses on the behavioural level. We show that both three-factor models and time series models can be used to uncover such time-dependencies. However, we argue that for short longitudinal data the three factor modelling approach is more suitable. In order to illustrate both approaches, we re-analysed previously published short longitudinal data sets. We found that in human embryonic kidney 293 cells cells the interaction effect in the regulation of extracellular signal-regulated kinase (ERK) 1 signalling activation by insulin and epidermal growth factor is subjected to a time effect and dramatically decays at peak values of ERK activation. In contrast, we found that the interaction effect induced by hypoxia and tumour necrosis factor-alpha for the transcriptional activity of the human cyclo-oxygenase-2 promoter in HEK293 cells is time invariant at least in the first 12-h time window after stimulation. Furthermore, we applied the three-factor model to previously reported animal studies. In these studies, memory storage was found to be subjected to an interaction effect of the beta-adrenoceptor agonist clenbuterol and certain antagonists acting on the alpha-1-adrenoceptor / glucocorticoid-receptor system. Our model-based analysis suggests that only if the antagonist drug is administer in a critical time window, then the interaction effect is relevant.

A comparison of sales response predictions from demand models applied to store-level versus panel data

In order to generate sales promotion response predictions, marketing analysts estimate demand models using either disaggregated (consumer-level) or aggregated (store-level) scanner data. Comparison of predictions from these demand models is complicated by the fact that models may accommodate different forms of consumer heterogeneity depending on the level of data aggregation. This study shows via simulation that demand models with various heterogeneity specifications do not produce more accurate sales response predictions than a homogeneous demand model applied to store-level data, with one major exception: a random coefficients model designed to capture within-store heterogeneity using store-level data produced significantly more accurate sales response predictions (as well as better fit) compared to other model specifications. An empirical application to the paper towel product category adds additional insights. This article has supplementary material online.

Role of fluctuations in membrane models:thermal versus nonthermal

We study the comparative importance of thermal to nonthermal fluctuations for membrane-based models in the linear regime. Our results, both in 1+1 and 2+1 dimensions, suggest that nonthermal fluctuations dominate thermal ones only when the relaxation time τ is large. For moderate to small values of τ, the dynamics is defined by a competition between these two forces. The results are expected to act as a quantitative benchmark for biological modeling in systems involving cytoskeletal and other nonthermal fluctuations. © 2011 American Physical Society.