Coefficients of agreement for fixed observers

Agreement between fixed observers or methods that produce readings on a continuous scale is usually evaluated via one of several intraclass correlation coefficients (ICCs). This article presents and discusses a few related issues that have not been raised before. ICCs are usually presented in the context of a two-way analysis of variance (ANOVA) model. We argue that the ANOVA model makes inadequate assumptions, such as the homogeneity of the error variances and of the pairwise correlation coefficients between observers. We then present the concept of observer relational agreement which has been used in the social sciences to derive the common ICCs without making the restrictive ANOVA assumptions. This concept did not receive much attention in the biomedical literature. When observer agreement is defined in terms of the difference of the readings of different observers on the same subject (absolute agreement), the corresponding relational agreement coefficient coincides with the concordance correlation coefficient (CCC), which is also an ICC. The CCC, which has gained popularity over the past 15 years, compares the mean squared difference between readings of observers on the same subject with the expected value of this quantity under the assumption of 'chance agreement', which is defined as independence between observers. We argue that the assumption of independence is unrealistic in this context and present a new coefficient that is not based on the concept of chance agreement.     

A general approach to evaluating agreement between two observers or methods of measurement from quantitative data with replicated measurements

We present a general approach to the definition and estimation of coefficients for evaluating agreement between two fixed methods of measurements or human observers. The measured variable is assumed to be continuous with a finite second moment. No other distributional assumptions are made. We introduce the term ;disagreement function' for the function of the observations that is used to quantify the extent of disagreement between the two measurements made on the same subject. The proposed inter-methods agreement coefficients compare the disagreement between measurements made by different methods on the same subject to the corresponding disagreement between replicated measurements made by the same method. Therefore, the new coefficients require data with replications readings. We propose inter-methods agreement coefficients for two practical situations involving two methods that have a measurement error: 1) comparison of a new method to a gold standard (or a reference method), and 2) comparison of two methods where neither method is considered a gold standard. We consider three disagreement functions based on the differences between two measurements: 1) the mean squared difference, 2) the mean absolute difference and 3) the mean relative difference. We then derive non-parametric estimates for the various agreement coefficients. Our approach is illustrated using data from a study comparing systolic blood pressure measurements by a human observer and an automatic monitor. The performance of the new estimates is assessed via stochastic simulations.     

THE ESTIMATION OF RANDOM COEFFICIENT AUTOREGRESSIVE MODELS. I

Abstract. This paper is concerned with autoregressive models in which the coefficients are assumed to be not constant but subject to random perturbations so that we are considering a class of random coefficient autoregressive models. By means of a two stage regression procedure estimates of the unknown parameters of these models are obtained. The estimates are shown to be strongly consistent and to satisfy a central limit theorem. A number of Monte Carlo experiments was carried out to illustrate the estimation procedure and their results are reported.     

缔合电解质溶液的积分方程理论

在粘性电解质模型（SEM）的基础上,研究了不等直径离子间有缔合作用的电解质溶液,其缔合位置小于或等于硬球的接触直径.求解Omstein-Zemike积分方程时,在核内和核外分别采用超网链近似（HNC）和平均球近似（MSA）作为封闭条件．系统的相关函数和热力学性质均表示为系统状态所决定的特性参数,并给出广泛的数值结果．     

Multidimensional scaling of selected samples from the optical society of America uniform color scales

The Optical Society of America Uniform Color Scales was produced as an attempt to create a color order system in which equal geometric distances correspond to equally perceived color differences. A psychophysical experiment was designed to determine the perceived visual uniformity within two oblique planar subsets of the Uniform Color Scales. A modified interval-scaling procedure using perceived value differences within a portion of the Munsell Value scale was used to assess the perceived magnitudes of dissimilarities between sample pairs of the Uniform Color Scales. Multidimensional-scaling methods were used to analyze the dissimilarity data for the two planar subsets, each of which was judged by five observers. Despite some limited subadditivity for which a number of plausible explanations were provided, each oblique planar subset was shown to demonstrate satisfactory visual uniformity on the basis of the average judgement obtained for five observers. These results also demonstrate that it is possible for human observers to judge dissimilarities within a planar array varying in three attributes of color and to resolve those differences into a true two-dimensional representation.     

A LINEAR ESTIMATION PROCEDURE FOR THE PARAMETERS OF AUTOREGRESSIVE MOVING-AVERAGE PROCESSES

Abstract. A linear estimation procedure for the parameters of autoregressive moving-average processes is proposed. The basic idea is to write the spectrum for the moving-average part as a linear function of a properly selected set of parameters and to use Chiu's weighted least-squares procedure to reduce the problem to a weighted linear least-squares problem. The proposed procedure finds estimates by solving systems of linear equations and does not need optimization programs. An one-step estimate is also suggested. It is shown that the estimates are asymptotically equal to the commonly used 'approximate' maximum likelihood estimate described in the paper. For Gaussian processes, the estimates obtained by the proposed procedures are asymptotically efficient.     

Ultrasonic characterization of the interface between a die attach adhesive and a copper leadframe in IC packaging

The silicon die (chip) and the copper leadframe in IC packaging are bonded by a die attach adhesive, and the quality of the interface is a critical issue in the reliability of IC packaging as well as during the manufacturing process. The common defects such as cracks and delamination can be detected using the C-mode scanning acoustic microscopy. However, a weak interface due to poor adhesion often goes undetected and may become a potential defective area at a later stage. This paper describes the work done to evaluate the quality of the weak interface between a die attach adhesive and a copper leadframe. An interface spring model is used to predict the ultrasonic reflection coefficients. Normal incidence reflection coefficients are measured from a two-layer specimen bonded with a die attach adhesive. The quality of the interface in the samples and its degradation are affected by copper oxidation, and by applying shear stress loading. It is shown that the reflection coefficient depends strongly on both the interface quality and stress loading, indicating that a nondestructive characterization of the interface is possible and the reflection coefficient can be used as a criterion.     

Estimation of true moments from truncated data

Moment analysis is a powerful tool in reactor bed analysis. Weighted moments is a technique which has been introduced to increase the accuracy of moment estimation. The supposed advantage of weighted moments is reevaluated utilizing an optimum truncation point. The resulting coefficient of variation in the moment estimates is shown to be superior to an arbitrary truncation point for all weighting factors. When an optimum truncation point is used, little advantage is observed for weighted moments.     

葡萄糖水溶液扩散系数的测定与关联

液相扩散系数在化工数据中起着非常重要的作用,由于液体结构及扩散本身的复杂性,至今还没有一个合理的理论模型来预测液相扩散系数,有必要从实验和理论上对不同的体系进行研究.制备了金属膜池,用氯化钾溶液标定了膜池常数,并用已有可靠文献值的蔗糖水溶液验证了装置的可靠性.测定了298.15～328.15 K不同温度下葡萄糖水溶液的积分扩散系数,考察了温度对扩散系数的影响.在实验数据的基础上,提出一个有温度关联项的半经验模型,关联和计算了本实验的数据,结果与实验值吻合很好.     

THE ESTIMATION OF RANDOM COEFFICIENT AUTOGRESSIVE MODELS. II

Abstract. In Nicholls and Quinn (1980) a procedure was proposed for the determination of strongly consistent estimates of random coefficient autoregressive models. These estimates are used here as starting values in a Newton-Raphson algorithm which is employed to obtain the maximum likelihood estimates of a class of random coefficient autoregressions. The maximum likelihood estimates are shown to be strongly consistent and to satisfy a central limit theorem. The problem of testing for the randomness of the coefficients is also briefly discussed. The results of a number of simulations are reported which illustrate the theoretical results obtained.     

Heat transfer metrology issues in two‐phase bubble column reactors

The metrology and the impact of various parameters and operating conditions on the bulk‐to‐tube heat transfer coefficients in two‐phase bubble columns are investigated on a small‐scale mock‐up. It is shown that (1) quasi‐adiabatic conditions can be reached in the column; (2) the bulk‐to‐tube heat transfer coefficients for each U‐tube downward and upward sections may or may not differ significantly, depending on the way uncertainty of the measurements is estimated; (3) using the different measurements and uncertainty estimates for given conditions, a mean heat transfer coefficient over all tubes is estimated within ±5%. The consequences for bulk‐to‐tube heat transfer coefficient prediction in a larger column are discussed.     

邻二甲苯在醋酸中扩散系数的测定

利用金属膜池并用氯化钾水溶液标定了膜池常数。用已有可靠文献值的乙酸水溶液验证了装置的可靠性。测定了298．15～323．15K不同温度下邻二甲苯在醋酸中的积分扩散系数,还分别给出了微分扩散系数与邻二甲苯浓度的关联式。结果表明在相同温度下邻二甲苯在醋酸中的扩散系数随着邻二甲苯浓度的增加而减小,在同一浓度下随着温度的升高而增加。     

Thermal Expansion of Orthorhombic PbO

Thermal expansion of orthorhombic PbO was investigated by high-temperature X-ray diffraction. The coefficients in the a ₀, and c ₀ directions are equal and <½ the coefficient in the b ₀ direction. The structure is described in terms of Pb–O chains extended in the a ₀ direction and bonded into layers in the b ₀ direction; appreciable Pb–Pb bonding is indicated. Coordination and bond distances observed in PbSiO₃ and lead silicate glasses, which are similar to those of massicot, and Pb _n O _n polymeric units reported in PbO vapor are interpreted as additional evidence of the importance of PbO chains as structural entities.     

对二甲苯在醋酸中扩散系数的测定

为了提供精对苯二甲酸（PTA）生产过程中相关体系的扩散系数,利用金属膜池并用氯化钾水溶液标定了膜池常数。用已有文献值的乙酸水溶液验证了装置的可靠性。测定了298.15-323.15K不同温度下对二甲苯在醋酸中的积分扩散系数,还分别给出了微分扩散系数与对二甲苯浓度的关联式。结果表明在相同温度下对二甲苯在醋酸中的扩散系数随着对二甲苯浓度的增加而减小,在同一浓度下随着温度的升高而增加。     

A pore network model for diffusion in nanoporous carbons: Validation by molecular dynamics simulation

A hybrid molecular dynamics simulation/pore network model (MD/PNM) approach is developed for predicting diffusion in nanoporous carbons. This approach is computationally fast, and related to the structure of the real material. The PNM takes into account both the geometrical (a distribution of pore sizes) and topological (the pore network connectivity) characteristics of nanoporous carbons, which are obtained by analysing adsorption data. The effective diffusion coefficient is calculated by taking the transport diffusion coefficients in single slit-shaped model pores from MD simulation and then computing the effective value over the PNM. The reliability of this approach is evaluated by comparing the results of the PNM analysis with a more rigorous, but much slower, simulation applied to a realistic model material, the virtual porous carbon (VPC). We obtain good agreement between the diffusion coefficients for the PNM and the VPC, indicating the reliability of the hybrid MD/PNM method and it can be used in industry for materials design.     

Evaluation of the reliability of anthropometric measures

As part of the Early Malnutrition and Its Effects on Youth Project, a study of the reliability of anthropometric measures was carried out through the re-measurement of 226 adolescents of the rural area of Guatemala. In all anthropometric variables, the intra-measure coefficients of reliability were higher than 0.96 and the inter-measure coefficients of reliability higher than 0.91. A significant effect of the measure was found, suggesting the existence of systematic differences among measures. This information will permit corrections by measure effect in the analysis of data. The effect of different grades of "data editing" and "validity, reliability and agreement check", of measures was evaluated. The conclusion was that even though the persons in charge of measurements were trained and supervised, the deletion of values outside ranks did not result in an appreciable decrease of the technical error of measurement. As expected, data editing and validity carried out through the comparison of two measurements taken from each subject and eliminating those whose differences were classified as error according to statistical criteria, decreased the technical error of measurement, in most of the variables measured. Nevertheless, with the exception of one of the variables, this decrease was not marked. This indicates that the technical error of measurement obtained from all the subjects, many of whom were not re-measured, was within acceptable ranks. Results of the study demonstrate the importance of including re-measurement of a certain percentage of subjects when designing anthropometric studies. This will allow evaluation of the reliability of measurements, identification of bias due to measurements, and their correction.     

Experimental investigation of axial and radial thermal dispersion in a packed bed

Axial and radial thermal dispersion coefficients are estimated under nonreacting and reacting conditions in an almost adiabatic packed bed using steady state temperature profiles and dynamic data simultaneously.Under nonreacting conditions it is concluded that in order to obtain reliable estimates of axial thermal dispersion coefficients it is necessary to incUnder reacting conditions the parameter estimates obtained for the thermal residence time and the axial thermal dispersion coefficient differ significa     

How temporal cues can aid colour constancy

Colour constancy assessed by asymmetric simultaneous colour matching usually reveals limited levels of performance in the unadapted eye. Yet observers can readily discriminate illuminant changes on a scene from changes in the spectral reflectances of the surfaces making up the scene. This ability is probably based on judgements of relational colour constancy, in turn based on the physical stability of spatial ratios of cone excitations under illuminant changes. Evidence is presented suggesting that the ability to detect violations in relational colour constancy depends on temporal transient cues. Because colour constancy and relational colour constancy are closely connected, it should be possible to improve estimates of colour constancy by introducing similar transient cues into the matching task. To test this hypothesis, an experiment was performed in which observers made surface-colour matches between patterns presented in the same position in an alternating sequence with period 2 s or, as a control, presented simultaneously, side-by-side. The degree of constancy was significantly higher for sequential presentation, reaching 87% for matches averaged over 20 observers. Temporal cues may offer a useful source of information for making colour-constancy judgements.     

Separative Diffusion in the Transient State. I. The Plate

Abstract

For a single plate the separation factor 5 depends only on the dimension-less time τ = Dt/Z ² and the difference between the diffusion coefficients ΔD. The output for a certain S can be increased linearly with the plate thickness while simultaneously the diffusion time increases with the square of the plate thickness. For a cascade of 2 plates of equal thickness, the time scale for equal S is expanded by a factor 4 against the single plate. Simultaneously the output is increased by a factor 4β, where β is the efficiency coefficient for the transfer between the 2 plates.     

Computer Algebra Derivation of the Bias of Linear Estimators of Autoregressive Models

Abstract. A symbolic method which can be used to obtain the asymptotic bias and variance coefficients to order O(1/n) for estimators in stationary time series is discussed. Using this method, the large‐sample bias of the Burg estimator in the AR(p) for p = 1, 2, 3 is shown to be equal to that of the least squares estimators in both the known and unknown mean cases. Previous researchers have only been able to obtain simulation results for the Burg estimator's bias because this problem is too intractable without using computer algebra. The asymptotic bias coefficient to O(1/n) of Yule–Walker as well as least squares estimates is also derived in AR(3) models. Our asymptotic results show that for the AR(3), just as in the AR(2), the Yule–Walker estimates have a large bias when the parameters are near the nonstationary boundary. The least squares and Burg estimates are much better in this situation. Simulation results confirm our findings.