Effect Of Sampling Error Research Articles

Allometry and phylogenetic divergence: Correspondence or incongruence?

The potential connection between trends of within species variation, such as those of allometric change in morphology, and phylogenetic divergence has been a central topic in evolutionary biology for more than a century, including in the context of human evolution. In this study, I focus on size-related shape change in craniofacial proportions using a sample of more than 3200 adult Old World monkeys belonging to 78 species, of which 2942 specimens of 51 species are selected for the analysis. Using geometric morphometrics, I assess whether the divergence in the direction of static allometries increases in relation to phyletic differences. Because both small samples and taxonomic sampling may bias the results, I explore the sensitivity of the main analyses to the inclusion of more or less taxa depending on the choice of a threshold for the minimum sample size of a species. To better understand the impact of sampling error, I also use randomized subsampling experiments in the largest species samples. The study shows that static allometries vary broadly in directions without any evident phylogenetic signal. This variation is much larger than previously found in ontogenetic trajectories of Old World monkeys, but the conclusion of no congruence with phylogenetic divergence is the same. Yet, the effect of sampling error clearly contributes to inaccuracies and tends to magnify the differences in allometric change. Thus, morphometric research at the boundary between micro- and macro-evolution in primates, and more generally in mammals, critically needs very large and representative samples. Besides sampling error, I suggest other non-mutually exclusive explanations for the lack of correspondence between allometric and phylogenetic divergence in Old World monkeys, and also discuss why directions might be more variable in static compared to ontogenetic trajectories. Even if allometric variation may be a poor source of information in relation to phylogeny, the evolution of allometry is a fascinating subject and the study of size-related shape changes remains a fundamental piece of the puzzle to understand morphological variation within and between species in primates and other animals.

Aeolian Sediment Transport Responses to Vegetation Cover Change: Effects of Sampling Error on Model Uncertainty

AbstractAlthough it is widely known that observations of aeolian sediment transport are susceptible to large sampling errors, sample designs are frequently used that do not sufficiently reduce the measurement uncertainties inherent in the study of aeolian processes. Here, we examine the influence of sample size (n) and sampling location on uncertainty in models of aeolian sediment transport responses to vegetation cover change. We compare measurements from a stratified random array of 27 horizontal sediment mass flux samplers to vegetative cover data collected at a 1 ha site over a period of nearly 6 years. To assess the sensitivity of modeled relationships between aeolian transport and vegetative cover to sample design, we analyze statistical regressions for all possible combinations of sample size and sampler locations. We show that at least 17 randomly located samplers are needed to consistently capture the sediment mass flux response to vegetative cover change. We found that multiple statistically significant models can describe the sediment flux‐vegetative cover relationship when using smaller sample sizes, demonstrating the risks of inferring sediment transport response from an underpowered sample design. Across vegetative functional groups, we found that woody cover generally influenced aeolian sediment transport rates more than herbaceous cover, while model uncertainty at large sample sizes (n > 17) showed the limitation of using vegetative cover as an indicator of aeolian sediment transport. Our results suggest an evaluation of sampling practices in aeolian sediment transport studies may be needed to avoid inferential errors that are likely pervasive in this field of study.

Effects of misrepresentative length samples on individual growth and stock condition estimates

Despite its importance in fisheries studies, there is insufficient understanding on the effect of sampling error or bias on individual growth and other stock indicators. We show the influence of sample length distributions on parameter estimates, illustrating with an example. For the brown swimming crab, we simulated length samples in five configurations and estimated parameters of von Bertalanffy (k, L∞L∞ , t0), asymptotic weight ( W∞W∞ ), weight-length relationship (a, b), growth performance (ϕ’) and condition factor (Kn). Parameter estimates were compared with baseline values using relative bias, standard error and root mean square error. The results show that the accuracy and bias of parameter estimates depend on the lengths sampled. For example, the bias and accuracy of L∞L∞ and W∞W∞ vary inversely with sampled length, whereas combining length segments yields smaller biases of k and t0 than those of L∞L∞ and W∞W∞ . In general, the accuracy of parameter estimates does not always depend on sampling the entire length range, and errors are not the same for all parameters. These results are useful to guide sampling when resources are scarce. We discuss potential reasons for incomplete length sample structure and offer recommendations to obtain best estimates for parameters of interest.

Framing TOS in risk assessment: an outreach perspective for the future

Sampling is necessary every time inferences are to be made to take informed, optimal decisions in science, technology, industry, trade and commerce. For reasons extensively addressed over the last two decades, some fields, normally those where good sampling practices are a source of economic gain such as the mining/minerals/metals industrial sectors, explicate the role of sampling more than others. This is not the case within the realm of food and feed safety assessment where sampling continues - still today - to be perceived more as an economic burden and a technical necessity to be fulfilled because of regulatory demands, rather than a need to ensure reliable evidence to support management and regulatory decisions. This is true today and will become even more central in the future to address the challenges posed by the accelerating climate crisis, resources depletion and increasing food demand. Risk assessment and sampling are both probabilistic disciplines, the first devoted to estimate and minimise safety risks, the latter devoted to estimate and mitigate sampling risks (the effects of sampling errors). Here we offer an exposé with the aim of positioning TOS as an essential disciplines and practical tool needed to ensure the best possible estimation of risks in support of safety decision-making and risk management in all of food and feed sciences, technology, industry, trade, commerce, and society at large. We demonstrate that sampling plays an integral, but an often much overlooked role in all these fields.

The Pearls and Perils of Google Trends: A Housing Market Application.

This study aims at providing insights into the correct usage of Google search data, which are available through Google Trends. The focus is on the effect of sampling errors, which has not received the attention that it deserves. A housing market application is used to demonstrate the effects. For this purpose, the relationship between online search activity for mortgages and real housing market activity is investigated. A simple time series model, which explains transactions by an online mortgage search, is estimated. The results show that the effect of sampling errors is substantial. Thus, although the application of Google Trends data in research remains promising, far more attention should be given to the limitations of these data.

Comparison of Conventional and Constrained Variational Methods for Computing Large‐Scale Budgets and Forcing Fields

AbstractAnalyses of atmospheric heat and moisture budgets serve as an effective tool to study convective characteristics over a region and to provide large‐scale forcing fields for various modeling applications. This paper examines two popular methods for computing large‐scale atmospheric budgets: the conventional budget method (CBM) using objectively gridded analyses based primarily on radiosonde data and the constrained variational analysis (CVA) approach which supplements vertical profiles of atmospheric fields with measurements at the top of the atmosphere and at the surface to conserve mass, water, energy, and momentum. Successful budget computations are dependent on accurate sampling and analyses of the thermodynamic state of the atmosphere and the divergence field associated with convection and the large‐scale circulation that influences it. Utilizing analyses generated from data taken during Dynamics of the Madden‐Julian Oscillation (DYNAMO) field campaign conducted over the central Indian Ocean from October to December 2011, we evaluate the merits of these budget approaches and examine their limitations. While many of the shortcomings of the CBM, in particular effects of sampling errors in sounding data, are effectively minimized with CVA, accurate large‐scale diagnostics in CVA are dependent on reliable background fields and rainfall constraints. For the DYNAMO analyses examined, the operational model fields used as the CVA background state provided wind fields that accurately resolved the vertical structure of convection in the vicinity of Gan Island. However, biases in the model thermodynamic fields were somewhat amplified in CVA resulting in a convective environment much weaker than observed.

A local sigma-point unscented Kalman filter for geophysical data assimilation

The prior covariance calculated in the Reduced-rank Sigma-Point Kalman filter (RSPKF) data assimilation method can be suboptimal as a result of finite number of sigma points, effects of sampling error, process error, and other factors. In this study, we analyse the performance of RSPKF by applying a localization scheme that combines local analysis and global generation of sigma points. The analysis at each model grid for the entire domain are updated using only the observations local to the analysis grid. The localization enables a high rank approximation of the background error covariance in a local subspace of greatly lower dimension than the global domain using a small number of sigma points. The global-analysis vector is constructed combining the local analyses at all model grid points. The global analysis-covariance matrix is generated in the ensemble subspace and the global sigma points are constructed following the RSPKF algorithm. Numerical experiments of our method utilized the Lorenz-96 model. The performance of the localization scheme is assessed in the presence of varying parameters such as the number of sigma points (k), inflation factor (ϕ), localization radius (d) and the number of model variables (Nx). When the localization is implemented, the number of sigma points required to achieve the minimum RMSE is significantly reduced compared to a case where no localization is used, for three different cases of model variables (Nx). We also show that the approximate number of sigma points used to obtain optimal estimate is independent of the state dimension Nx of the model. This further highlights the importance of localization in RSPKF, making it a potential candidate for data assimilation in oceanic or atmospheric General Circulation Models (GCMs).

The Molecular Microscope (MMDXR ) interpretation of thoracic and abdominal allograft biopsies: Putting things in perspective for the clinician.

The Molecular Microscope System (MMDXR ) has significant potential to enhance biopsy interpretation. However, discussions of MMDx do not acknowledge the basic accuracy of histology readings, and the ability of pathology as a stand-alone tool to guide patient management. MMDx overstates its ability to automatically correct for problems in biopsy readings. Assertions of accuracy approaching 99% are not supported by "real world" data. The high rate of discrepancies between MMDx® and standard biopsy readings can be attributed to the summation of many factors other than histology interpretation, including molecular noise, assay thresholding, limited sensitivity of microarray technology for low expression genes, errors in classifier development, narrow data interpretation, and lack of spatial context. It is not widely recognized that molecular signatures are not disease-specific and are affected by the stage of disease as well as the extent of tissue injury. The effect of sampling error on MMDx performance is significantly under-estimated, particularly in heart and lung biopsies. Therefore, MMDx reports should always be interpreted in the context of conventional biopsy readings. The clinical picture and conventional biopsy reading should be allowed to over-ride the molecular interpretation when there is concern that confounding factors are at play.

Effect of core needle biopsy number on intraductal carcinoma diagnosis in patients with metastatic castration-sensitive prostate cancer.

243 Background: The number of core needle biopsies performed to diagnose metastatic prostate cancer are sometimes reduced in daily clinical practice. Intraductal carcinoma of the prostate (IDC-P) is associated with adverse prognostic parameters and has recently received attention as a valuable indicator for the prediction of disease severity. Currently, the relationship between IDC-P diagnosis and the number of core prostate biopsies is unclear. In the present study, we analyzed the effect of core needle biopsy number on IDC-P diagnosis in patients with metastatic castration-sensitive prostate cancer. Methods: We retrospectively evaluated data from 150 patients diagnosed with metastatic prostate cancer at our hospital between 2002 and 2012. Percentage of the core involved with cancer and the maximum cancer occupancy were 100% in median of all the patients. Subjects were allocated to three groups according to the number of core biopsies performed: ≤5, 6–9, and ≥10. The study endpoints were the cancer-specific survival (CSS) and overall survival (OS) rates. Results: Twenty-seven (18%) patients had ≤5 core biopsies, 67 (45%) patients had 6–9 core biopsies, and 56 (37%) patients had ≥10 core biopsies. For patients who underwent ≥10 core biopsies, a significant difference on CSS was detected between with or without IDC-P ( p = 0.002). On the other hand, the difference decreased as the number of core biopsies became smaller (6–9; p = 0.322 and ≤5; p = 0.815). A similar trend was identified for the OS outcome. A significant difference on OS was also found between with or without IDC-P in patients who underwent ≥10 and 6–9 core needle biopsies ( p = 0.0002 and 0.015, respectively), but not in those who underwent ≤5 core biopsies ( p = 0.3407). IDC-P served as a stronger prognostic marker for CSS and OS than did the other factors included in the multivariate analysis. ( p = 0.0024, and p = 0.0014, respectively). Conclusions: Given the IDC-P detection and its value as a prognostic marker, we recommend the performance of ≥10 core biopsy procedures in patients diagnosed with metastatic prostate cancer. Sampling error effects the IDC-P value in smaller number of core needle biopsies.

Metodi di campionamento negli studi epidemiologici

Sampling methods in epidemiological studies Sampling allows researchers to obtain information about a population through data obtained from a subset of the population, with a saving in terms of costs and workload compared to a study based on the entire population. Sampling allows the collecting of high quality information, provided that the sample size is large enough to detect a true association between exposure and outcome. There are two types of sampling methods: probability and non-probability sampling. In probability sampling the subset of the population is extracted randomly from all eligible individuals; this method, as all subjects have a chance of being chosen, allows researchers to generalize the findings of their study. In non-probability sampling, some individuals have no chance of being selected, because researchers do not extract the sample from all eligible subjects of a population; the sample is probably non-representative, the effect of sampling error cannot be estimated, so that the study produces non-generalizable results. Examples of probability sampling methods are: simple random sampling, systematic sampling, stratified sampling, and clustered sampling. Examples of non-probability sampling methods are: convenience sampling, judgement sampling.

Assimilating the along-track sea level anomaly into the regional ocean modeling system using the ensemble optimal interpolation

The ensemble optimal interpolation (EnOI) is applied to the regional ocean modeling system (ROMS) with the ability to assimilate the along-track sea level anomaly (TSLA). This system is tested with an eddy-resolving system of the South China Sea (SCS). Background errors are derived from a running seasonal ensemble to account for the seasonal variability within the SCS. A fifth-order localization function with a 250 km localization radius is chosen to reduce the negative effects of sampling errors. The data assimilation system is tested from January 2004 to December 2006. The results show that the root mean square deviation (RMSD) of the sea level anomaly decreased from 10.57 to 6.70 cm, which represents a 36.6% reduction of error. The data assimilation reduces error for temperature within the upper 800 m and for salinity within the upper 200 m, although error degrades slightly at deeper depths. Surface currents are in better agreement with trajectories of surface drifters after data assimilation. The variance of sea level improves significantly in terms of both the amplitude and position of the strong and weak variance regions after assimilating TSLA. Results with AGE error (AGE) perform better than no AGE error (NoAGE) when considering the improvements of the temperature and the salinity. Furthermore, reasons for the extremely strong variability in the northern SCS in high resolution models are investigated. The results demonstrate that the strong variability of sea level in the high resolution model is caused by an extremely strong Kuroshio intrusion. Therefore, it is demonstrated that it is necessary to assimilate the TSLA in order to better simulate the SCS with high resolution models.

The Effects of Sampling Errors on the EnKF Assimilation of Inner-Core Hurricane Observations

Abstract Atmospheric data assimilation methods that estimate flow-dependent forecast statistics from ensembles are sensitive to sampling errors. This sensitivity is investigated in the context of vortex-scale hurricane data assimilation by cycling an ensemble Kalman filter to assimilate observations with a convection-permitting mesoscale model. In a set of numerical experiments, airborne Doppler radar observations are assimilated for Hurricane Katrina (2005) using an ensemble size that ranges from 30 to 300 members, and a varying degree of covariance inflation through relaxation to the prior. The range of ensemble sizes is shown to produce variations in posterior storm structure that persist for days in deterministic forecasts, with the most substantial differences appearing in the vortex outer-core wind and pressure fields. Ensembles with 60 or more members converge toward similar axisymmetric and asymmetric inner-core solutions by the end of the cycling, while producing qualitatively similar sample correlations between the state variables. Though covariance relaxation has little impact on model variables far from the observations, the structure of the inner-core vortex can benefit from a more optimal tuning of the relaxation coefficient. Results from this study provide insight into how sampling errors may affect the performance of an ensemble hurricane data assimilation system during cycling.

Satellite precipitation in southeastern South America: how do sampling errors impact high flow simulations?

Satellite precipitation estimates are increasingly available at temporal and spatial scales of interest to hydrological applications and with the potential for improving flood forecasts in data-sparse regions. This study evaluates the effect of sampling error on simulated large flood events. Synthetic precipitation fields were generated in Monte Carlo fashion by perturbing observed precipitation fields with sampling errors based on 1, 2 and 6 h intervals. The variable infiltration capacity hydrological model was used to assess the impact of these errors on simulated high flow events in the Iguazu basin, a rain-dominated, subtropical basin in southeastern South America. Results showed that unbiased errors in daily error-corrupted precipitation fields introduced bias in the simulated hydrologic fluxes and states. The overall bias for error-corrupted daily streamflows was positive and its magnitude increased with larger sampling intervals. However, for high flow events, the bias was negative as a result of an increase in simulated infiltration and changes in precipitation variability. Errors in precipitation also affected the magnitude and volume of the peak events but did not change the first two statistical moments of the peaks indicating that non-linearities in the hydrological system preserve the statistical properties of high flows in the basin. Caution is needed when using satellite products for hydrological applications that require the estimation of large peaks and volumes.

Dynamic analysis of Bayesian audit strategies with tests of controls and reliability modeling

PurposeThe purpose of this paper is to summarize a simulation study that analyzed the performance of Bayesian audit strategies in a novel fashion – dynamically and with varying sample sizes depending on the extent of an auditor's prior information.Design/methodology/approachThe prior information for the Bayesian strategies arises from a set of control tests that are evaluated making use of reliability theory. The entire audit strategy is simulated under systematically different control reliabilities and related amounts of total misstatements in an accounting population.FindingsThe major finding is that robust Bayesian audit strategies that have recently been developed in auditing research are more sensitive to non‐sampling errors than existing strategies of audit practice.Practical implicationsThe authors find that there are differential effects of sampling error vs non‐sampling error on the Bayesian strategies and that controls testing does not need to be extensive to get full internal control reliance.Originality/valueThe paper adds to existing research by examining the performance of various Bayesian audit strategies under more realistic audit conditions of sampling and non‐sampling uncertainty.

Improved initial ensemble generation coupled with ensemble square root filters and inflation to estimate uncertainty

The performance of the Ensemble Kalman Filter method (EnKF) depends on the sample size compared to the dimension of the parameters space. In real applications insufficient sampling may result in spurious correlations which reduce the accuracy of the filter with a strong underestimation of the uncertainty. Covariance localization and inflation are common solutions to these problems. The Ensemble Square Root Filters (ESRF) is also better to estimate uncertainty with respect to the EnKF. In this work we propose a method that limits the consequences of sampling errors by means of a convenient generation of the initial ensemble. This regeneration is based on a Stationary Orthogonal-Base Representation (SOBR) obtained via a singular value decomposition of a stationary covariance matrix estimated from the ensemble. The technique is tested on a 2D single phase reservoir and compared with the other common techniques. The evaluation is based on a reference solution obtained with a very large ensemble (one million members) which remove the spurious correlations. The example gives evidence that the SOBR technique is a valid alternative to reduce the effect of sampling error. In addition, when the SOBR method is applied in combination with the ESRF and inflation, it gives the best performance in terms of uncertainty estimation and oil production forecast.

Visualisation of Sampling Error Effects in near Infrared Analysis—Comparison between Petri Dish, Roll Bottle and Spiral Sampler

With spectroscopic methods, e.g. near infrared (NIR) analysis, using a constant beam aperture, the effective scanning footprint will be different for a spinning Petri dish, a rolling bottle and a new spiral sampler configuration. This will significantly influence the analytical accuracy and precision of a NIR analytical determination of heterogeneous materials, for example barley with differing protein contents. Here we present the results from a bench-top experiment that evaluates the total analytical bias and precision characteristics for three alternative sample presentation approaches using a mixture of two plastic polymer pellets as a test material with significant heterogeneity. After removal of all incorrect sampling errors (ICS), there are still significantly varying correct sampling error [Fundamental Sampling Error ( FSE) and Grouping and Segregation Error ( GSE)] uncertainties associated with these standard analytical approaches–but there is a clear winner.

Sampling Errors in Seasonal Forecasting

Abstract The limited numbers of start dates and ensemble sizes in seasonal forecasts lead to sampling errors in predictions. Defining the magnitude of these sampling errors would be useful for end users as well as informing decisions on resource allocation to minimize total system error. A numerical experiment has been designed to measure them, and results indicate that sampling errors are substantial in state-of-the-art seasonal forecast systems. The standard solution of increasing sample sizes is of limited benefit in seasonal forecasting because of restrictions imposed by resource costs and nonstationary observations. Alternative options, based on the postprocessing of forecast and hindcast data, are presented in this paper. The spatial and temporal aggregations of data together with the appropriate use of theoretical distributions can reduce the effect of sampling errors on forecast quantities by an amount equivalent to increasing samples sizes by a factor of 4 of more, with insignificant losses of forecast information. These postprocessing techniques can be viewed as cost-effective methods of reducing the effects of sampling errors in seasonal forecast quantities.

Effects of sampling error and temporal correlations in population growth on process variance estimators

Estimates of a population’s growth rate and process variance from time-series data are often used to calculate risk metrics such as the probability of quasi-extinction, but temporal correlations in the data from sampling error, intrinsic population factors, or environmental conditions can bias process variance estimators and detrimentally affect risk predictions. It has been claimed (McNamara and Harding, Ecol Lett 7:16–20, 2004) that estimates of the long-term variance that incorporate observed temporal correlations in population growth are unaffected by sampling error; however, no estimation procedures were proposed for time-series data. We develop a suite of such long-term variance estimators, and use simulated data with temporally autocorrelated population growth and sampling error to evaluate their performance. In some cases, we get nearly unbiased long-term variance estimates despite ignoring sampling error, but the utility of these estimators is questionable because of large estimation uncertainty and difficulties in estimating correlation structure in practice. Process variance estimators that ignored temporal correlations generally gave more precise estimates of the variability in population growth and of the probability of quasi-extinction. We also found that the estimation of probability of quasi-extinction was greatly improved when quasi-extinction thresholds were set relatively close to population levels. Because of precision concerns, we recommend using simple models for risk estimates despite potential biases, and limiting inference to quantifying relative risk; e.g., changes in risk over time for a single population or comparative risk among populations.

Effects of sampling errors on abundance estimates from virtual population analysis for walleye pollock in northern waters of Sea of Japan

Effects of sampling errors on abundance estimates from virtual population analysis (VPA) were quantified with the bootstrap method for stock of walleye pollock in the the Sea of Japan. In the bootstrap method, individual fish measurements were resampled. A total of 1000 bootstrap samples were produced for each year from 1991 to 2001. The coefficients of variation (CV) of catch at age in 2001 ranged 6.1–33.1%. The CV of an abundance estimate in 2001 ranged 9.0–35.7%. Abundance estimates of the oldest age and the latest year, which had larger CVs than the other estimates, were sensitive to sampling errors. Effects of sampling errors became smaller when the catch at age had been accumulated over a few years. Although VPA includes various types of errors, only the sampling errors have room for improvement in reality. Quantifying the effect of sampling error on VPA estimates is essential for sound and efficient stock management, and is emphasized in this study.

Effects of sampling error on bioassessments of stream ecosystems: application to RIVPACS-type models

AbstractWe evaluated the influence of various sources of sampling error on the precision, accuracy, and sensitivity of bioassessments based on River Invertebrate Prediction and Classification System (RIVPACS)-type models. We used data from 98 minimally altered streams in western Oregon and Washington to generate 18 models representing 2 field collection techniques (fixed-area riffle and multiple-habitat collections) and 9 levels of subsampling (50, 100, …450 fixed counts). For each model, we generated 2 observed-to-expected taxa ratios (O/E). The 1st O/E was based on all predicted taxa and the 2nd excluded rare taxa (i.e., those taxa with predicted probabilities of occurrence <0.5). We then compared O/E values to determine the extent to which subsampling effort, field collection method, different field personnel, and individual site characteristics altered model performance. We also generated O/E values for 63 streams with varying amounts of watershed and channel alteration (test sites) to quantify the ex...