Estimation Of Various Parameters Research Articles

Personality, plasticity and predictability in sticklebacks: bold fish are less plastic and more predictable than shy fish

It is well established that animals often differ consistently from one another in their behaviour. Most work has focused on consistent differences in average behaviour, generally referred to as ‘animal personality’. However, individuals may also differ consistently from one another in how they change their behaviour over time or across environments, that is, plasticity, as well as in the predictability of their behaviour. How these three sources of behavioural variation are linked is rarely explored. Here we tested 80 wild-caught three-spined sticklebacks, Gasterosteus aculeatus, repeatedly over a 10-week period for their tendency to explore an open arena from the safety of cover, generally termed ‘boldness’. Using a mixed-modelling approach, we detected consistent individual differences in boldness, as well as in the extent to which fish changed this behaviour over time, that is, temporal plasticity. Boldness and plasticity were negatively correlated, with shy fish spending increasing amounts of time out of cover while bold fish changed little over time. After accounting for variation in personality and plasticity, we still detected individual differences in intraindividual behavioural variation (‘IIV’), that is, individual predictability. This residual variation was linked to fish’ boldness score, with shy fish having higher IIV and thus being less predictable in their behaviour than bold fish. After a 4-week break from testing, individuals on average decreased their time out of cover again to baseline levels. However, shyer, more plastic fish showed the largest changes in behaviour. Statistical simulations show that our experimental data set has sufficient statistical power to provide accurate and precise variance parameter estimates. Together, our results provide strong evidence that individual variation in boldness is linked with behavioural plasticity and predictability. Establishing the evolutionary causes as well as fitness consequences of these links in behavioural variation is an exciting challenge for future research.

Solution of the Ill-Posed Semiparametric Regression Model Based on Singular Value Modification Restriction

We propose a solution of the ill-posed semi-parametric regression model based on singular value modification restriction, aimed at the ill-posed problem of the normal matrix which may occur in the process of solving the semiparametric regression model. First, the coefficient matrix is decomposed into singular values, and the smaller singular values are selected according to the criterion ∑i=1r1/σi/∑i=1n1/σi≤5% (in the singular value matrix, σ1>σ1>⋯>σr>⋯σn). Second, the relatively smaller singular values are modified by the biased parameter to suppress the magnification of the estimated variance so as to effectively reduce the variance of parameter estimation, reduce the introduction of deviation and obtain more reliable parameter estimation. The results of the numerical experiments show that the improved singular value modification restriction method can not only overcome the effect of the ill-posed normal matrix on the parameter estimation solution but also correctly separate the systematic errors and improve the accuracy of semiparametric regression model calculation results.

Estimation of Variance for Reciprocal General and Specific Combining Ability Effects by EM‐AI Algorithm

ABSTRACTIn diallel analysis, using mixed models, it is possible to estimate the variance components of general combining ability (GCA), specific combining ability (SCA), reciprocal GCA (RGCA), reciprocal SCA (RSCA), and parents, even in unbalanced datasets. These variances can be estimated by likelihood maximization via numerical algorithms (e.g., expectation maximization [EM] and average information [AI]), which have different advantages. Thus, the objective of this study was to describe and implement the EM‐AI algorithm (combination of EM and AI) in R software to estimate the variance components of RGCA, RSCA, and parents effects in diallel analysis. Two real datasets and three diallel models (Griffing's Model 1, Griffing's Model 1 + RGCA, and Model 3, a general diallel model with the effects of GCA, SCA, RGCA, RSCA, and parents) were used to evaluate the efficiency of the algorithms AI, EM, and EM‐AI. The AI algorithm failed to converge for the three diallel models in both datasets. The other algorithms (EM and EM‐AI) converged normally, and the estimated variance components with these algorithms were similar for the three diallel models in both datasets. However, the EM‐AI algorithm was more efficient than the EM algorithm, and the general diallel model (Model 3) provided more accurate estimates of variance parameters. Thus, the EM‐AI algorithm with routine implemented in R has potential for use in diallel analyses in plant breeding.

Efficient Monte Carlo algorithm for restricted maximum likelihood estimation of genetic parameters.

Monte Carlo (MC) methods have been found useful in estimation of variance parameters for large data and complex models with many variance components (VC), with respect to both computer memory and computing time. A disadvantage has been a fluctuation in round-to-round values of estimates that makes the estimation of convergence challenging. Furthermore, with Newton-type algorithms, the approximate Hessian matrix might have sufficient accuracy, but the inaccuracy in the gradient vector exaggerates the round-to-round fluctuation to intolerable. In this study, the reuse of the same random numbers within each MC sample was used to remove the MC fluctuation. Simulated data with six VC parameters were analysed by four different MC REML methods: expectation-maximization (EM), Newton-Raphson (NR), average information (AI) and Broyden's method (BM). In addition, field data with 96 VC parameters were analysed by MC EM REML. In all the analyses with reused samples, the MC fluctuations disappeared, but the final estimates by the MC REML methods differed from the analytically calculated values more than expected especially when the number of MC samples was small. The difference depended on the random numbers generated, and based on repeated MC AI REML analyses, the VC estimates were on average non-biased. The advantage of reusing MC samples is more apparent in the NR-type algorithms. Smooth convergence opens the possibility to use the fast converging Newton-type algorithms. However, a disadvantage from reusing MC samples is a possible "bias" in the estimates. To attain acceptable accuracy, sufficient number of MC samples need to be generated.

Average information residual maximum likelihood in practice.

Gilmour, Thompson, and Cullis (Biometrics, 1995, 51, 1440) presented the average information residual maximum likelihood (REML) algorithm for efficient variance parameter estimation in the linear mixed model. That paper dealt specifically with traditional variance component models, but the algorithm was quickly applied to more general models and implemented in several REML packages including ASReml (Gilmour etal., Biometrics, 2015, 51, 1440). This paper outlines the theory with respect to these more general models, describes the main issues encountered in fitting these models and how they have been addressed in the ASReml software. The issues covered are the basics steps in the implementation of the algorithm, keeping parameters within the parameter space, maximizing sparsity, avoiding issues associated with unstructured variance matrices by using the factor-analytic structure and handling singularities in marker-based relationship matrices and current work.

Effects of globally obtained informative priors on bayesian safety performance functions developed for Australian crash data

The precision and bias of Safety Performance Functions (SPFs) heavily rely on the data upon which they are estimated. When local (spatially and temporally representative) data are not sufficiently available, the estimated parameters in SPFs are likely to be biased and inefficient. Estimating SPFs using Bayesian inference may moderate the effects of local data insufficiency in that local data can be combined with prior information obtained from other parts of the world to incorporate additional evidence into the SPFs. In past applications of Bayesian models, non-informative priors have routinely been used because incorporating prior information in SPFs is not straightforward. The previous few attempts to employ informative priors in estimating SPFs are mostly based on local prior knowledge and assuming normally distributed priors. Moreover, the unobserved heterogeneity in local data has not been taken into account. As such, the effects of globally derived informative priors on the precision and bias of locally developed SPFs are essentially unknown.This study aims to examine the effects of globally informative priors and their distribution types on the precision and bias of SPFs developed for Australian crash data. To formulate and develop global informative priors, the means and variances of parameter estimates from previous research were critically reviewed. Informative priors were generated using three methods: 1) distribution fitting, 2) endogenous specification of dispersion parameters, and 3) hypothetically increasing the strength of priors obtained from distribution fitting. In so doing, the mean effects of crash contributing factors across the world are significantly different than those same effects in Australia. A total of 25 Bayesian Random Parameters Negative Binomial SPFs were estimated for different types of informative priors across five sample sizes. The means and standard deviations of posterior parameter estimates as well as SPFs goodness of fit were compared between the models across different sample sizes. Globally informative prior for the dispersion parameter substantially increases the precision of a local estimate, even when the variance of local data likelihood is small. In comparison with the conventional use of Normal distribution, Logistic, Weibull and Lognormal distributions yield more accurate parameter estimates for average annual daily traffic, segment length and number of lanes, particularly when sample size is relatively small.

Do Poorer Children Have Poorer Playgrounds? A Geographically Weighted Analysis of Attractiveness, Cleanliness, and Safety of Playgrounds in Affluent and Deprived Urban Neighborhoods

Background: Playgrounds are a central resource when it comes to physical activity among minors. This study assesses the association between area deprivation and the quality of playgrounds. Methods: All playgrounds in the city of Mannheim, Germany (145km2, 311,000 inhabitants) were visited between July 2016 and January 2017 as part of a systematic audit. Each playground's amenities, attractiveness, cleanliness, and safety were operationalized using well-established, validated instruments. Global and geographically weighted regression models were fitted to investigate the association of the amenities, attractively, cleanliness, and safety of playgrounds with sociodemographic indicators on the social area level. Results: A total of 271 playgrounds were identified. Overall, population density showed the strongest association with all quality variables in the global models, followed by the central official poverty indicator. Significant spatial variation in parameter estimates was found for most of the deprivation indicators with regard to attractiveness, cleanliness, and safety of playgrounds indicating locally negative associations between area-level deprivation and quality. Conclusion: Our findings illustrate the importance of a qualitative approach by analyzing physical activity resources. Concerning the quality of playgrounds, the data from this study support the deprivation amplification hypothesis, meaning that children who are already socially disadvantaged might experience a further contextual disadvantage.

Dielectric behavior and structural characterization of polymeric double layer thin films

ABSTRACTIn the present work, various techniques have been employed to investigate the polarization processes along with the estimation of various parameters involved therein in polymethylmethacrylate (PMMA), polyvinylidenefluoride (PVDF), and PVDF‐PMMA double‐layered samples. Anticipating the difficulty to analyze the experimental data; in such complicated situation, when in one the same temperature range various decay mechanism may complete, the different techniques are used to interpret the experimental results. These measurements verify the various processes responsible polarization in these polymer samples. The structural morphology, crystallinity, atomic, and molecular geometry were analyzed using different characterization tools, namely SEM, XRD, FTIR, and DSC; therefore, the results were found to be in confirmation with amorphous nature of PMMA and semicrystalline nature of PVDF and PVDF‐PMMA double‐layered samples. Ac dielectric measurements were performed for different temperatures over the frequency range of 100 Hz to 10 kHz. Dielectric losses have also been recorded with a two terminal (Al‐Al) electrode combination. The study revealed that both interfacial phenomena and space charge mechanism are known to play preliminary role in the dielectric behavior of double‐layered samples. The amorphous–crystalline interface act as charge‐carrier trapping sites in this polymeric heterogeneous system. Dielectric permittivity and dielectric loss of the double‐layered sample depend manifolds on the temperature and the Ac frequency. © 2019 Wiley Periodicals, Inc. J. Appl. Polym. Sci. 2019, 136, 47724.

Probabilistic Characterization of Rock Mass from Limited Laboratory Tests and Field Data: Associated Reliability Analysis and Its Interpretation

Probabilistic methods are the most efficient methods to account for different types of uncertainties encountered in the estimated rock properties required for the stability analysis of rock slopes and tunnels. These methods require estimation of various parameters of probability distributions like mean, standard deviation (SD) and distributions types of rock properties, which requires large amount of data from laboratory and field investigations. However, in rock mechanics, the data available on rock properties for a project are often limited since the extents of projects are usually large and the test data are minimal due to cost constraints. Due to the unavailability of adequate test data, parameters (mean and SD) of probability distributions of rock properties themselves contain uncertainties. Since traditional reliability analysis uses these uncertain parameters (mean and SD) of probability distributions of rock properties, they may give incorrect estimation of the reliability of rock slope stability. This paper presents a method to overcome this limitation of traditional reliability analysis and outlines a new approach of rock mass characterization for the cases with limited data. This approach uses Sobol’s global sensitivity analysis and bootstrap method coupled with augmented radial basis function based response surface. This method is capable of handling the uncertainties in the parameters (mean and SD) of probability distributions of rock properties and can include their effect in the stability estimates of rock slopes. The proposed method is more practical and efficient, since it considers uncertainty in the statistical parameters of most commonly and easily available rock properties, i.e. uniaxial compressive strength and Geological Strength Index. Further, computational effort involved in the reliability analysis of rock slopes of large dimensions is comparatively smaller in this method. Present study also demonstrates this method through reliability analysis of a large rock slope of an open pit gold mine in Karnataka region of India. Results are compared with the results from traditional reliability analysis to highlight the advantages of the proposed method. It is observed that uncertainties in probability distribution type and its parameters (mean and SD) of rock properties have considerable effect on the estimated reliability index of the rock slope and hence traditional reliability methods based on the parameters of probability distributions estimated using limited data can make incorrect estimation of rock slope stability. Further, stability of the rock slope determined from proposed approach based on bootstrap method is represented by confidence interval of reliability index instead of a fixed value of reliability index as in traditional methods, providing more realistic estimates of rock slope stability.

Prediction of Subgenome Additive and Interaction Effects in Allohexaploid Wheat.

Whole genome duplications have played an important role in the evolution of angiosperms. These events often occur through hybridization between closely related species, resulting in an allopolyploid with multiple subgenomes. With the availability of affordable genotyping and a reference genome to locate markers, breeders of allopolyploids now have the opportunity to manipulate subgenomes independently. This also presents a unique opportunity to investigate epistatic interactions between homeologous orthologs across subgenomes. We present a statistical framework for partitioning genetic variance to the subgenomes of an allopolyploid, predicting breeding values for each subgenome, and determining the importance of inter-genomic epistasis. We demonstrate using an allohexaploid wheat breeding population evaluated in Ithaca, NY and an important wheat dataset from CIMMYT previously shown to demonstrate non-additive genetic variance. Subgenome covariance matrices were constructed and used to calculate subgenome interaction covariance matrices for variance component estimation and genomic prediction. We propose a method to extract population structure from all subgenomes at once before covariances are calculated to reduce collinearity between subgenome estimates. Variance parameter estimation was shown to be reliable for additive subgenome effects, but was less reliable for subgenome interaction components. Predictive ability was equivalent to current genomic prediction methods. Including only inter-genomic interactions resulted in the same increase in accuracy as modeling all pairwise marker interactions. Thus, we provide a new tool for breeders of allopolyploid crops to characterize the genetic architecture of existing populations, determine breeding goals, and develop new strategies for selection of additive effects and fixation of inter-genomic epistasis.

Factor Analysis of Employee Motivation towards 5S Culture

The 5S system is a system of elimination, tidiness, cleanliness, continuous standardization and habitual behavior. Implementing the 5S system requires employee motivation so that the application of 5S becomes effective and sustainable. From year 2015 to year 2017, the performance of 5S system implementation at an automotive manufacturer had been fluctuate. In year 2020, the company has a mission to achieve Total Productive Maintenance Award which is prestigious in manufacturing industry. The results of the study show that there is a direct or positive influence between employee motivation towards the culture of 5S system implementation. This can explain that employee motivation has a positive effect on the 5S culture. The parameter estimation value of the highest employee motivation variable is on the Improve variable indicator which is considered to have a direct effect. The result indicates that the employee's needs for career development and information are met in the company. The highest value of cultural variable parameter estimation of 5S system implementation is on the Shine variable indicator which is considered to have a direct effect. The result indicates that most employees can apply cleaning practice and cleaning maintenance to goods and work areas compared to other 5S system steps.

The WIG (weighted individual and group) shrinkage estimator.

Regularization, or shrinkage estimation, refers to a class of statistical methods that constrain the variability of parameter estimates when fitting models to data. These constraints move parameters toward a group mean or toward a fixed point (e.g., 0). Regularization has gained popularity across many fields for its ability to increase predictive power over classical techniques. However, articles published in JEAB and other behavioral journals have yet to adopt these methods. This paper reviews some common regularization schemes and speculates as to why articles published in JEAB do not use them. In response, we propose our own shrinkage estimator that avoids some of the possible objections associated with the reviewed regularization methods. Our estimator works by mixing weighted individual and group (WIG) data rather than by constraining parameters. We test this method on a problem of model selection. Specifically, we conduct a simulation study on the selection of matching-law-based punishment models, comparing WIG with ordinary least squares (OLS) regression, and find that, on average, WIG outperforms OLS in this context.

Genetic divergence, traits association, path analysis and harvest index in pigeonpea (Cajanus Cajan L.)

Twenty six genotypes of pigeonpea were evaluated for 13 agro-morphological characters. Statistical analysis was carried out for estimation of various parameters. The ICPL-87 being shortest genotypes possessed the minimum number of branches, minimum pod weight, lowest 100-seed weight, shortest root length, lowest biological yield per plant and minimum yield per plant. Among all the genotypes, NDA-1 was the tallest genotype. The highest harvest index was showed by NDA-2009-1. Considering the performance of all the genotypes for important characters, including yield, the genotypes; NDA-2009-1, PUSA-992, DA-10-2, MAL-6, ICPL-87 and BRG-2 were promising. Root weight showed very high estimate of genetic advance (85.53%). The value of genetic advance was lowest for root length (20.89%). The residual effect (0.064) indicated that the thirteen characters included in this study explain 93.6 per cent of variation in yield in this population. Plant height, pod weight, root weight and harvest index were positively associated with seed yield simultaneously with positive direct effect. Therefore, direct selection against this character would be effective for seed yield improvement in pigeonpea.

Use of likelihood ratio tests to detect outliers under the variance shift outlier model

ABSTRACTIn this paper, we revisit the alternative outlier model of Thompson [A note on restricted maximum likelihood estimation with an alternative outlier model, J. Roy. Stat. Soc. Ser. B 47 (1985), pp. 53–55] for detecting outliers in the linear model. Gumedze et al. [A variance shift model for detection of outliers in the linear mixed model, Comput. Statist. Data Anal. 54 (2010), pp. 2128–2144] called this model the variance shift outlier model (VSOM). The basic idea behind the VSOM is to detect observations with inflated variance and isolate them for further investigation. The VSOM is appealing because it downweights an outlier in the analysis, with the weighting determined automatically as part of the estimation procedure. We set up the VSOM as a linear mixed model and then use the likelihood ratio test (LRT) statistic as an objective measure for determining whether the weighting is required, i.e. whether the observation is an outlier. We also derived one-step updates of the variance parameter estimates based on observed, expected and average information matrices to obtain one-step LRT statistics which usually require less computation. Both the fully iterated and one-step LRTs are functions of the squared standard residuals from the null model and therefore can be computed directly without the need to fit the VSOM. We investigated the properties of the likelihood ratio tests and compare them. An extension of the model to detect a group of outliers is also given. We illustrate the proposed methodology using simulated datasets and a real dataset.

Lamotrigine pharmacokinetics following oral and stable-labeled intravenous administration in young and elderly adult epilepsy patients: Effect of age.

The objectives of this study were to investigate the effect of age on pharmacokinetic parameters of lamotrigine (LTG) and estimate parameter variability. Patients (>18years old) who were already on a steady-state dose of LTG therapy with no interacting comedications were enrolled. Patients with significant cardiac disease, severe kidney dysfunction, or moderate-to-severe liver dysfunction were excluded. Fifty milligrams of a stable-labeled intravenous LTG formulation (SL-LTG) replaced 50mg of a patient's normal daily oral LTG dose. Thirteen blood samples were collected in each person over 96hours. SL-LTG and unlabeled LTG concentrations were measured simultaneously by gas chromatography-mass spectrometry. Concentration-time data were analyzed by nonlinear mixed-effects modeling (NONMEM version 7.3). Twenty-eight patients representing 16 young (18-48years old) and 12 elderly (63-87years old) patients were included, yielding 382 unlabeled and 351 SL-LTG concentrations. A two-compartment model with first-order absorption and elimination adequately described the plasma concentration-time data. Bioavailability of oral LTG was approximately 74% and did not differ by age. LTG clearance was 27.2% lower in elderly than in young patients (1.80L/h for a 70-kg patient). Although LTG bioavailability was not affected by age, LTG clearance was 27.2% lower in elderly versus young patients of comparable body weight, possibly indicating lower dosages being needed in this population.

Phenotypic and Genotypic Trends for Body Weights Traits in Romanov Sheep

Data on 820 Romanov lambs progeny of 30 sires it covered the period from 1995 to 2005 were used in this study. Lambs traits studied were birth weight (BW) body weight at one month (BW1), body weight at two months (BW2), weaning weight at three month (WW) and average daily gain from birth to weaning (ADG). Data were analysis by using mixed model. Means of BW, Bw1, BW2, WW and ADG were 2.90 kg, 7.10 kg, 10.50 kg, 13.02 kg and 115.50 g respectively. Sire of lambs, ewes within sires had a highly significant effect on all traits. Season and year of lambing, type of birth and sex had a significant effect on all studied traits, expect the effect 0f year lambing on BW and type of birth on BW1 and WW. Also, inbreeding coefficients had a highly significant effect on all body weight traits studied and decreased as inbreeding coefficient increased. Two animal models were used. Model 1 includes the fixed effects season and year of lambing, type of birth and sex and random effects of direct genetic effect, permanent environmental effect and residual effect. Model 2, is similar to model 1 and added maternal genetic effect and covariance between direct and maternal genetic effect.Determination of direct heritability for body weights traits ranged from 0.17 to 0.39 for model 1 and ranged from 0.13 to 0.29 for model 2. The removal of additive maternal effects and covariance between direct and maternal genetic effects (model 1) increased estimates of direct heritability. Therefore, including maternal effects the model resulted in more accurate estimation of (co) variance and genetic parameters of growth traits. Determination of phenotypic and genetic correlations among growth traits studied were moral and highly significant. While, annual phenotypic and genetic trends for body weights traits were negative.

On the estimation of variance parameters in non-standard generalised linear mixed models: application to penalised smoothing

We present a novel method for the estimation of variance parameters in generalised linear mixed models. The method has its roots in Harville (1977)'s work, but it is able to deal with models that have a precision matrix for the random-effect vector that is linear in the inverse of the variance parameters (i.e., the precision parameters). We call the method SOP (Separation of Overlapping Precision matrices). SOP is based on applying the method of successive approximations to easy-to-compute estimate updates of the variance parameters. These estimate updates have an appealing form: they are the ratio of a (weighted) sum of squares to a quantity related to effective degrees of freedom. We provide the sufficient and necessary conditions for these estimates to be strictly positive. An important application field of SOP is penalised regression estimation of models where multiple quadratic penalties act on the same regression coefficients. We discuss in detail two of those models: penalised splines for locally adaptive smoothness and for hierarchical curve data. Several data examples in these settings are presented.

Does an uneven sample size distribution across settings matter in cross-classified multilevel modeling? Results of a simulation study

BackgroundRecent advances in multilevel modeling allow for modeling non-hierarchical levels (e.g., youth in non-nested schools and neighborhoods) using cross-classified multilevel models (CCMM). Current practice is to cluster samples from one context (e.g., schools) and utilize the observations however they are distributed from the second context (e.g., neighborhoods). However, it is unknown whether an uneven distribution of sample size across these contexts leads to incorrect estimates of random effects in CCMMs. MethodsUsing the school and neighborhood data structure in Add Health, we examined the effect of neighborhood sample size imbalance on the estimation of variance parameters in models predicting BMI. We differentially assigned students from a given school to neighborhoods within that school's catchment area using three scenarios of (im)balance. 1000 random datasets were simulated for each of five combinations of school- and neighborhood-level variance and imbalance scenarios, for a total of 15,000 simulated data sets. For each simulation, we calculated 95% CIs for the variance parameters to determine whether the true simulated variance fell within the interval. ResultsAcross all simulations, the “true” school and neighborhood variance parameters were estimated 93–96% of the time. Only 5% of models failed to capture neighborhood variance; 6% failed to capture school variance. ConclusionsThese results suggest that there is no systematic bias in the ability of CCMM to capture the true variance parameters regardless of the distribution of students across neighborhoods. Ongoing efforts to use CCMM are warranted and can proceed without concern for the sample imbalance across contexts.

On the suitability of an allometric proxy for nondestructive estimation of average leaf dry weight in eelgrass shoots I: sensitivity analysis and examination of the influences of data quality, analysis method, and sample size on precision

BackgroundThe effects of current anthropogenic influences on eelgrass (Zostera marina) meadows are noticeable. Eelgrass ecological services grant important benefits for mankind. Preservation of eelgrass meadows include several transplantation methods. Evaluation of establishing success relies on the estimation of standing stock and productivity. Average leaf biomass in shoots is a fundamental component of standing stock. Existing methods of leaf biomass measurement are destructive and time consuming. These assessments could alter shoot density in developing transplants. Allometric methods offer convenient indirect assessments of individual leaf biomass. Aggregation of single leaf projections produce surrogates for average leaf biomass in shoots. Involved parameters are time invariant, then derived proxies yield simplified nondestructive approximations. On spite of time invariance local factors induce relative variability of parameter estimates. This influences accuracy of surrogates. And factors like analysis method, sample size and data quality also impact precision. Besides, scaling projections are sensitive to parameter fluctuation. Thus the suitability of the addressed allometric approximations requires clarification.ResultsThe considered proxies produced accurate indirect assessments of observed values. Only parameter estimates fitted from raw data using nonlinear regression, produced robust approximations. Data quality influenced sensitivity and sample size for an optimal precision.ConclusionsAllometric surrogates of average leaf biomass in eelgrass shoots offer convenient nondestructive assessments. But analysis method and sample size can influence accuracy in a direct manner. Standardized routines for data quality are crucial on granting cost-effectiveness of the method.

Automated threshold selection for extreme value analysis via ordered goodness-of-fit tests with adjustment for false discovery rate

Threshold selection is a critical issue for extreme value analysis with threshold-based approaches. Under suitable conditions, exceedances over a high threshold have been shown to follow the generalized Pareto distribution (GPD) asymptotically. In practice, however, the threshold must be chosen. If the chosen threshold is too low, the GPD approximation may not hold and bias can occur. If the threshold is chosen too high, reduced sample size increases the variance of parameter estimates. To process batch analyses, commonly used selection methods such as graphical diagnosis are subjective and cannot be automated, while computational methods may not be feasible. We propose to test a set of thresholds through the goodness-of-fit of the GPD for the exceedances, and select the lowest one, above which the data provides adequate fit to the GPD. Previous attempts in this setting are not valid due to the special feature that the multiple tests are done in an ordered fashion. We apply two recently available stopping rules that control the false discovery rate or familywise error rate to ordered goodness-of-fit tests to automate threshold selection. Various model specification tests such as the Cramer-von Mises, Anderson-Darling, Moran's, and a score test are investigated. The performance of the method is assessed in a large scale simulation study that mimics practical return level estimation. This procedure was repeated at hundreds of sites in the western US to generate return level maps of extreme precipitation.