Model-free model elimination: A new step in the model-free dynamic analysis of NMR relaxation data

Model-free analysis is a technique commonly used within the field of NMR spectroscopy to extract atomic resolution, interpretable dynamic information on multiple timescales from the R ₁, R ₂, and steady state NOE. Model-free approaches employ two disparate areas of data analysis, the discipline of mathematical optimisation, specifically the minimisation of a χ² function, and the statistical field of model selection. By searching through a large number of model-free minimisations, which were setup using synthetic relaxation data whereby the true underlying dynamics is known, certain model-free models have been identified to, at times, fail. This has been characterised as either the internal correlation times, τ _e , τ _f , or τ _s , or the global correlation time parameter, local τ _m , heading towards infinity, the result being that the final parameter values are far from the true values. In a number of cases the minimised χ² value of the failed model is significantly lower than that of all other models and, hence, will be the model which is chosen by model selection techniques. If these models are not removed prior to model selection the final model-free results could be far from the truth. By implementing a series of empirical rules involving inequalities these models can be specifically isolated and removed. Model-free analysis should therefore consist of three distinct steps: model-free minimisation, model-free model elimination, and finally model-free model selection. Failure has also been identified to affect the individual Monte Carlo simulations used within error analysis. Each simulation involves an independent randomised relaxation data set and model-free minimisation, thus simulations suffer from exactly the same types of failure as model-free models. Therefore, to prevent these outliers from causing a significant overestimation of the errors the failed Monte Carlo simulations need to be culled prior to calculating the parameter standard deviations.     

The Generalised Estimating Equations: A Comparison of Procedures Available in Commercial Statistical Software Packages

The Generalised Estimating Equations (GEE) proposed by Liang and Zeger (1986) and Zeger and Liang (1986) have found considerable attention in the last decade (for an overview see e.g. Ziegler, and Blettner , 1998). Several self-made programs for solving the GEE are available. This paper presents a comparison of three GEE procedures that are already available in SAS PROC GENMOD, STATA procedure XTGEE and SUDAAN PROC MULTILOG. We show that the estimation results may be quite distinct due to different implementations. Summing up, it is pleasant that GEE is becoming established in commercial software packages. However, some aspects of the implementations should be improved.     

Sensitivity analysis of an energetic muscle model applied at whole body level in recumbent pedalling

Musculoskeletal models are used in order to describe and analyse the mechanics of human movement. In order to get a complete evaluation of the human movement, energetic muscle models were developed and were shown to be promising.

The aim of this work is to determine the sensitivity of muscle mechanical and energetic model estimates to changes in parameters during recumbent pedalling.

Inputs of the model were electromyography and joint angles, collected experimentally on one participant. The sensitivity analysis was performed on muscle-specific tension, physiological cross-sectional area, muscle maximal force, tendon rest length and percentage of fast-twitch fibres using an integrated sensitivity ratio. Soleus, gastrocnemius, vasti, gluteus and medial hamstrings were selected for the analyses.

The energetic model was found to be always less sensitive to parameter changes than the mechanical model. Tendon slack length was found to be the most critical parameter for both energetic and mechanical models even if the effect on the energetic output was smaller than on muscle force and joint moments.     

生态模型的灵敏度分析

灵敏度分析用于定性或定量地评价模型参数误差对模型结果产生的影响,是模型参数化过程和模型校正过程中的有用工具,具有重要的生态学意义．灵敏度分析包括局部灵敏度分析和全局灵敏度分析．局部灵敏度分析只检验单个参数的变化对模型结果的影响程度;全局灵敏度分析则检验多个参数的变化对模型运行结果总的影响,并分析每一个参数及其参数之间相互作用对模型结果的影响．目前,在对生态模型的灵敏度分析中,越来越倾向于使用全局灵敏度分析的方法．但国内仍多采用局部灵敏度分析方法,很少采用全局灵敏度分析方法．文中详细论述了局部灵敏分析和全局灵敏度分析的主要方法(一次变换法、多元回归法、Morris法、Sobol’法、傅里叶幅度灵敏度检验法和傅里叶幅度灵敏度检验扩展法),希望能为国内生态模型的发展提供一个比较完善的灵敏度分析方法库．结合国内外的灵敏度分析发展现状,指出联合灵敏度研究、灵敏度共性研究及空间直观景观模型的灵敏度分析将为生态模型灵敏度分析研究中的热点和难点．     

Reliability Analysis of the Groundwater Conceptual Model

The hydrologic model is the foundation of water resource management and planning. Conceptual model is the essential component of groundwater model. Due to limited understanding of natural hydrogeological conditions, the conceptual model is always constructed incompletely. Therefore, the uncertainty in the model's output is evitable when natural groundwater field is simulated by a single groundwater model. A synthetic groundwater model is built and regarded as the true model, and three alternative conceptual models are constructed by considering incomplete hydrogeological conditions. The outputs (groundwater budget terms from boundary conditions) of these groundwater models are analyzed statistically. The results show that when the conceptual model is closer to the true hydrogeological conditions, the distributions of outputs of the groundwater model are more concentrated on the true outputs. Therefore, the more reliable the structure of the conceptual model is, the more reliable the output of the groundwater model is. Moreover, the uncertainty caused by the conceptual model cannot be compensated by parameter uncertainty.     

Evaluation of longevity modeling censored records in Nellore

The aim of the present study was to evaluate the prediction ability of models that cope with longevity phenotypic expression as uncensored and censored in Nellore cattle. Longevity was defined as the difference between the dates of last weaned calf and cow birth. There were information of 77 353 females, being 61 097 cows with uncensored phenotypic information and 16 256 cows with censored records. These data were analyzed considering three different models: (1) Gaussian linear model (LM), in which only uncensored records were considered; and two models that consider both uncensored and censored records: (2) Censored Gaussian linear model (CLM); and (3) Weibull frailty hazard model (WM). For the model prediction ability comparisons, the data set was randomly divided into training and validation sets, containing 80% and 20% of the records, respectively. There were considered 10 repetitions applying the following restrictions: (a) at least three animals per contemporary group in the training set; and (b) sires with more than 10 progenies with uncensored records (352 sires) should have daughters in the training and validation sets. The variance components estimated using the whole data set in each model were used as true values in the prediction of breeding values of the animals in the training set. The WM model showed the best prediction ability, providing the lowest χ² average and the highest number of sets in which a model had the smallest value of χ² statistics. The CLM and LM models showed prediction abilities 2.6% and 3.7% less efficient than WM, respectively. In addition, the accuracies of sire breeding values for LM and CLM were lower than those obtained for WM. The percentages of bulls in common, considering only 10% of sires with the highest breeding values, were around 75% and 54%, respectively, between LM–CLM and LM–WM models, considering all sires, and 75% between LM–CLM and LM–WM, when only sires with more than 10 progenies with uncensored records were taken into account. These results are indicative of reranking of animals in terms of genetic merit between LM, CLM and WM. The model in which censored records of longevity were excluded from the analysis showed the lowest prediction ability. The WM provides the best predictive performance, therefore this model would be recommended to perform genetic evaluation of longevity in this population.     

The Generalised Estimating Equations: An Annotated Bibliography

The Generalised Estimating Equations (GEE) proposed by Liang and Zeger (1986) and Zeger and Liang (1986) have found considerable attention in the last ten years and several extensions have been proposed. In this annotated bibliography we describe the development of the GEE and its extensions during the last decade. Additionally, we discuss advantages and disadvantages of the different parametrisations that have been proposed in the literature. Furthermore, we review regression diagnostic techniques and approaches for dealing with missing data. We give an insight to the different fields of application in biometry. We also describe the software available for the GEE.     

A periodic analysis of longitudinal binary responses: a case study of clinical mastitis in Norwegian Red cows

A Bayesian procedure for analyzing longitudinal binary responses using a periodic cosine function was developed. It was assumed that, after adjustment for "seasonal" effects, the oscillation of the underlying latent variables for longitudinal binary responses was a stationary series. Based on this assumption, a single dimension sinusoidal analysis of longitudinal binary responses using the Gibbs sampling and Metropolis algorithms was implemented in a study of clinical mastitis records of Norwegian Red cows taken over five lactations.     

生态环境系统结构模型解析及其应用

一、系统结构模型解析及其作用在生态环境研究中,经常遇到复杂的生态环境系统。为了把握这样的复杂系统的总体结构,我们一般用图这个数学工具为其建立模型。把系统所含的因素当作结点,因素两两之间的关系连成线,即可得到我们研究对象的系     

Robust semiparametric microarray normalization and significance analysis

Microarray technology allows the monitoring of expression levels of thousands of genes simultaneously. A semiparametric location and scale model is proposed to model gene expression levels for normalization and significance analysis purposes. Robust estimation based on weighted least absolute deviation regression and significance analysis based on the weighted bootstrap are investigated. The proposed approach naturally combines normalization and significance analysis, and incorporates the variations due to normalization into the significance analysis properly. A small simulation study is used to compare finite sample performance of the proposed approach with alternatives. We also demonstrate the proposed method with a real dataset.     

Panel estimation for the impacts of population-related factors on CO2 emissions: A regional analysis in China

A large accumulation of carbon dioxide emission have attracted much attention recently. The existing researches mainly focused on such impact factors of carbon dioxide emission as population, economy, technology and others. However, there is little specific guidance for the subdivision of demographic factors. This paper employed STIRPAT (Stochastic Impacts by Regression on Population, Affluence and Technology) model to examine the impact of population size, per capita consumption, energy intensity, urbanization and aging population on CO₂ emissions by adopting panel data of 30 provinces from 1997 to 2012. Taking the climate change as a control variable, we can get the result that the population size, per capita consumption and energy intensity have strong explanatory power on CO₂ emissions in the three regions. The urbanization level has a positive influence on carbon emissions in the western region and has a negative effect in the central region, while it is not statistically significant in the eastern region. Aging population increases emissions in the eastern region, while decreases emissions in the central region and the western region.     

Gene-based modelling for rice: an opportunity to enhance the simulation of rice growth and development?

Process-based crop simulation models require employment of new knowledge for continuous improvement. To simulate growth and development of different genotypes of a given crop, most models use empirical relationships or parameters defined as genetic coefficients to represent the various cultivar characteristics. Such a loose introduction of different cultivar characteristics can result in bias within a simulation, which could potentially integrate to a high simulation error at the end of the growing season when final yield at maturity is predicted. Recent advances in genetics and biomolecular analysis provide important opportunities for incorporating genetic information into process-based models to improve the accuracy of the simulation of growth and development and ultimately the final yield. This improvement is especially important for complex applications of models. For instance, the effect of the climate change on the crop growth processes in the context of natural climatic and soil variability and a large range of crop management options (e.g., N management) make it difficult to predict the potential impact of the climate change on the crop production. Quantification of the interaction of the environmental variables with the management factors requires fine tuning of the crop models to consider differences among different genotypes. In this paper we present this concept by reviewing the available knowledge of major genes and quantitative trait loci (QTLs) for important traits of rice for improvement of rice growth modelling and further requirements. It is our aim to review the assumption of the adequacy of the available knowledge of rice genes and QTL information to be introduced into the models. Although the rice genome sequence has been completed, the development of gene-based rice models still requires additional information than is currently unavailable. We conclude that a multidiscipline research project would be able to introduce this concept for practical applications.     

应用物元评判识别模型预测农业害虫种群动态

论述了物元评判识别模型的建模方法,并探讨了该模型在预测农业害虫种群动态方面的应用．结果表明,其历史拟合率高达 ９２．３１％,将 １９９５年作为独立样本进行试报,其预报结果与实际发生一致．并指出物元评判模型是预测农业害虫种群动态的一种优良模型．     

Mapping cladograms into morphospaces

In cladistic analyses, taxa are grouped hierarchically into clades according to shared apomorphic character states to construct cladograms; cladograms are interpretable as phylogenetic hypotheses. In morphological space analyses, organism forms are represented as points in morphospaces; point proximities in morphospaces represent similarities that might be attributable to phenetic convergence and, consequently, may correspond inaccurately with hypothesized evolutionary relationships. A method for synthesizing phylogenetic results that are interpreted from cladistic analyses with phenetic results that are obtained from morphological space analyses is presented here; in particular, points that represent forms typifying taxa in morphospace are assigned as terminal nodes for appropriate cladograms that are mapped into morphospaces by positioning nonterminal nodes and orienting internodes according to a geometric algorithm. Nonterminal nodes may be interpreted as ancestors in phylogenetic hypotheses and occupy positions that represent particular organism forms in morphospaces. By mapping cladograms into morphospaces, therefore, evolutionary morphologists can reconstruct ancestral morphologies and test historical transformation hypotheses.