Showing papers on "Sampling (statistics) published in 1988"

X charts with variable sampling intervals

Abstract: The usual practice in using a control chart to monitor a process is to take samples from the process with fixed sampling intervals. This article considers the properties of the chart when the sampling interval between each pair of samples is not fixed but rather depends on what is observed in the first sample. The idea is that the time interval until the next sample should be short if a sample shows some indication of a change in the process and long if there is no indication of a change. The proposed variable sampling interval (VSI) chart uses a short sampling interval if is close to but not actually outside the control limits and a long sampling interval if is close to target. If is actually outside the control limits, then the chart signals in the same way as the standard fixed sampling interval (FSI) chart. Properties such as the average time to signal and the average number of samples to signal are evaluated. Comparisons between the FSI and the VSI charts indicate that the VSI chart is substantially ...

Resampling Inference with Complex Survey Data

Abstract: Methods for standard errors and confidence intervals for nonlinear statistics —such as ratios, regression, and correlation coefficients—have been extensively studied for stratified multistage designs in which the clusters are sampled with replacement, in particular, the important special case of two sampled clusters per stratum. These methods include the customary linearization (or Taylor) method and resampling methods based on the jackknife and balanced repeated replication (BRR). Unlike the jackknife or the BRR, the linearization method is applicable to general sampling designs, but it involves a separate variance formula for each nonlinear statistic, thereby requiring additional programming efforts. Both the jackknife and the BRR use a single variance formula for all nonlinear statistics, but they are more computing-intensive. The resampling methods developed here retain these features of the jackknife and the BRR, yet permit extension to more complex designs involving sampling without replace...

Estimating Total Fish Abundance and Total Habitat Area in Small Streams Based on Visual Estimation Methods

Abstract: We present sampling designs for estimating total areas of habitat types and total fish numbers in small streams. Designs are applied independently within strata constructed on the basis of habitat ...

Improvement Of Second‐Order Reliability Estimates by Importance Sampling

Abstract: The calculation of probability integrals is one of the most important tasks in structural reliability. First-order and asymptotic second-order methods have been proposed. The method of importance sampling is applied to determine the remaining error and to define a correction factor for the asymptotic second-order methods. The method can be made arbitrarily exact at the expense of more numerical effort.

Calculating confidence intervals for some non-parametric analyses.

Abstract: Gardner and Altman1 described the rationale behind the use of confidence intervals and gave methods for their calculation for a population mean and for differences between two population means for paired and unpaired samples. These methods are based on sample means, standard errors, and the t distribution and should strictly be used only for continuous data from Normal distributions (although small deviations from Normality are not important2). For non-Normal continuous data the median of the population or the sample is preferable to the mean as a measure of location. Medians are also appropriate in other situations?for example, when measurements are on an ordinal scale. This paper describes methods of calculating confidence intervals for a population median or for other population quantiles from a sample of observations. Calculations of confidence intervals for the difference between two population medians or means (a non-parametric approach rather than the parametric approach mentioned above) for both unpaired and paired samples are described. Worked examples are given for each situation. Because of the discrete nature of some of the sampling distribu? tions involved in non-parametric analyses it is not usually possible to calculate confidence intervals with exactly the desired level of confidence. Hence, if a 95% confidence interval is wanted the choice is between the lowest possible level of confidence over 95% (a "conservative" interval) and the highest possible under 95%. There is no firm policy on which of these is preferred, but we will mainly describe conservative intervals in this paper. The exact level of confidence associated with any particular approximate level can be calculated from the distribution of the statistic being used. The methods outlined for obtaining confidence intervals are described in more detail in textbooks on non-parametric statistics.3 The calculations can be carried out using the statistical computer package MINITAB.4 A method for calculating confidence intervals for Spearman's rank correlation coefficient is given in an accom? panying paper.5 A confidence interval indicates the precision of the sample statistic as an estimate of the overall population value. Confidence intervals convey the effects of sampling variation but cannot control for non-sampling errors in study design or conduct. They should not be used for basic description of the sample data but only for indicating the uncertainty in sample estimates for population values of medians or other statistics.

Probability Integration by Directional Simulation

Abstract: Reliability evaluation methods based on directional simulation for structural reliability analysis is treated herein. The methods can be used to check results obtained by a first‐ or second‐order reliability method. Directional importance sampling based on first‐ and second‐order reliability results is used to reduce the variance on the probability estimator. Sampling densities and procedures for important types of failure surfaces considered in the space of independent and standardized Gaussian variables are set up. The types include hyper‐spheres, hyperplanes, rotationally symmetrical hyperparaboloids, surfaces defining a convex polyhedral failure set, as well as series‐systems failure surfaces.

Large aperture time-series sediment traps; design objectives, construction and application

Abstract: Sediment traps with 0.5 and 1.15 m2 apertures which are capable of collecting 12–25 samples at programmed intervals, typically weekly or bi-monthly, during one continuous semi- to interannual deployment have been developed. They utilize a number of new synthetic materials and stable metallic components which ensure reliable, long-lasting performance at any oceanic depth. The key component of the trap is a set of sequentially rotating samplers which is driven by a microprocessor-controlled electronic stepping motor. The electronic power controller controls sampler exchange with a high degree of flexibility and precision, as well as independently recording the executed sampling events. Each sampling bottle is sealed from ambient water during the time samples are stored before recovery. After continuous improvement and modification during 29.5 deployment-years of application in deep ocean experiments since 1982, we are convinced that these sediment traps can provide a relatively large quantity of settling particles in time-series with high experimental reliability.

Sampling rare and elusive populations.

Abstract: The sampling of rare and elusive populations is difficult because the costs of locating such populations are substantial and can exceed actual interviewing costs. There are efficient probability methods that have been developed recently that reduce these costs. If the special populations are geographically clustered, efficient sampling involves the rapid location of segments in which no members of the special population are located with the use of Census data, telephone screening, or incomplete lists. Populations that are not geographicaily clustered can be located by network sampling and use of large previously gathered samples. Characteristics of mobile populations such as the homeless can be estimated by capture-recapture methods.

Large Sample Theory of Empirical Distributions in Biased Sampling Models

Abstract: Vardi (1985a) introduced an $s$-sample model for biased sampling, gave conditions which guarantee the existence and uniqueness of the nonparametric maximum likelihood estimator $\mathbb{G}_n$ of the common underlying distribution $G$ and discussed numerical methods for calculating the estimator. Here we examine the large sample behavior of the NPMLE $\mathbb{G}_n$, including results on uniform consistency of $\mathbb{G}_n$, convergence of $\sqrt n (\mathbb{G}_n - G)$ to a Gaussian process and asymptotic efficiency of $\mathbb{G}_n$ as an estimator of $G$. The proofs are based upon recent results for empirical processes indexed by sets and functions and convexity arguments. We also give a careful proof of identifiability of the underlying distribution $G$ under connectedness of a certain graph $\mathbf{G}$. Examples and applications include length-biased sampling, stratified sampling, "enriched" stratified sampling, "choice-based" sampling in econometrics and "case-control" studies in biostatistics. A final section discusses design issues and further problems.

Improved importance sampling technique for efficient simulation of digital communication systems

Abstract: An improved importance sampling technique for the efficient simulation of digital communication systems is proposed. Evaluation of low-probability error events by direct use of classical Monte Carlo (MC) simulation techniques usually involves a very large number of runs. Importance-sampling techniques make the low-probability events occur more frequently. The technique proposed here is based on optimized translations of the original probability densities. The only approximation needed in the optimizations is that of replacing the Q function by the simpler exponential expression. Detailed analytical evaluations of the estimation variances of the classical MC and the conventional and improved importance-sampling approaches for systems with memories and signals are presented and compared, showing the superior performance of the latter. Detailed numerical and simulation results are given. >

Methodology for Obtaining a Representative Sample of Homeless Persons The Los Angeles Skid Row Study

Abstract: Recent efforts on the part of survey researchers to understand the characteristics and needs of homeless individuals have been hampered by factors which make it extra ordinarily difficult to draw representative samples of this population. To date, only one study (Rossi et al., 1987) has drawn a probability sample of homeless persons that includes unsheltered individuals. Because the design of the Rossi study can only accommodate a short interview and is best carried out in one night, additional designs that allow more lengthy interview protocols and data collection periods are needed. An effort to draw a probability sample of homeless adults in the inner-city area of Los Angeles resulted in a sampling design that meets these criteria. This article describes this design in detail.

Principles of Random Variate Generation

Abstract: Author's preface. Algorithmic conventions. Glossary. Simulation and random variate generation random number sequences general methods for generating random variates from unvariate distributions methods of generation from specific continuous distributions discrete distributions multivariate distributions miscellaneous topics, including the generation of order statistics, simulation of stochastic processes, sampling from distribution tails. Appendices: 1 - FORTRAN 77 routines for generating variates from selected distributions 2 - graphical methods for sampling from standardized gama distributions and the standardized normal distributions. References. Index.

Assessment of sampling stability in ecological applications of discriminant analysis

Abstract: A simulation study was undertaken to assess the sampling stability of the variable loadings in linear discriminant function analysis. A factorial design was used for the factors of multivariate dimensionality, dispersion structure, configuration of group means, and sample size. A total of 32 400 discriminant analyses were conducted, based on data from simulated populations with appropriate underlying statistical distributions. Results from the simulations suggest that minimum sample sizes must exceed multi- variate dimensionality by at least a factor of three to achieve reasonable levels of stability in discriminant function loadings. However, the requisite sample size would vary with respect to each of the design factors and, especially, with the overall amount of system variation. A review of 60 published studies and 142 individual analyses indicated that sample sizes in ecological studies often have met that requirement. However, individual group sample sizes frequently were very unequal, and checks of assumptions usually were not reported. We recommend that ecologists obtain group sample sizes that are at least three times as large as the number of variables measured.

Statistical estimators for relational algebra expressions

Abstract: Present database systems process all the data related to a query before giving out responses. As a result, the size of the data to be processed becomes excessive for real-time/time-constrained environments. A new methodology is needed to cut down systematically the time to process the data involved in processing the query. To this end, we propose to use data samples and construct an approximate synthetic response to a given query.In this paper, we consider only COUNT(E) type queries, where E is an arbitrary relational algebra expression. We make no assumptions about the distribution of attribute values and ordering of tuples in the input relations, and propose consistent and unbiased estimators for arbitrary COUNT(E) type queries. We design a sampling plan based on the cluster sampling method to improve the utilization of sampled data and to reduce the cost of sampling. We also evaluate the performance of the proposed estimators.

Automated blood sampling systems for positron emission tomography

Abstract: An automated blood sampling system has been constructed and evaluated for use with positron-emission tomography (PET) The dispersion functions of two different detector units in the blood sampling system are compared One is a plastic scintillator, and the other is a coincidence detector No significant differences were found Results from studies of blood-brain barrier transfer of a C-11 labelled receptor antagonist are discussed >

Ecology of fish spawning and nursery zones in the flood plain, using a new sampling approach

Abstract: Horizontal zonation of fish reproduction, a lotic-to-lentic succession similar to that seen with increasing stream order, was evident from the relative abundance of larval and 0 + juvenile fishes in three floodplain spawning and nursery areas (lotic, semi-lotic, lentic) of the Upper Rhone River, France. Although the lotic and lentic ecosystems provided similar estimates of standing crop (0 + juveniles), differences were apparent in the reproductive and trophic guild structure of the YOY taxocoenoses at the three sites. A new sampling approach (Random Point-Abundance Sampling and modified electrofishing) is described for early-life fish ecology. The electrofishing method employed is mobile, effective for all sizes of larvae and 0 + juveniles of most species, quantitative, and applicable to a number of freshwater situations; and the punctual data resulting from this sampling approach are comparable both spatially and temporally.

Comments on the testing of the accuracy of an extraction procedure for determining the partitioning of trace metals in sediments

Abstract: Discussion sur l'article de Kheboian, C. et Bauer, C.F., Anal. chem. 1987, vol. 59, 1417-1423

Bootstrap and other methods to measure errors in survey estimates

Abstract: Variance estimation techniques for nonlinear statistics, such as ratios and regression and correlation coefficients, and functionals, such as quantiles, are reviewed in the context of sampling from stratified populations. In particular, resampling methods such as the bootstrap, the jackknife, and balanced repeated replication are compared with the traditional linearization method for nonlinear statistics and a method based on Woodruff's confidence intervals for the quantiles. Results of empirical studies are presented on the bias and stability of these variance estimators and on confidence‐interval coverage probabilities and lengths. Copyright

The Cdf Central and Endwall Hadron Calorimeter

Abstract: The CDF central and endwall hadron calorimeter covers the polar region between 30° and 150° and a full 2π in azimuth. It consists of 48 steel-scintillator central modules with 2.5 cm sampling and 48 steel-scintillator endwall modules with 5.0 cm sampling. A general description of the detector is given. Calibration techniques and performance are discussed. Some results of the test beam studies are shown.

Sampling properties of estimators of the log‐logistic distribution with application to Canadian precipitation data

Abstract: We consider the probability-weighted moment and the maximum-likelihood estimators of two parameters in the log-logistic distribution. Quantile estimators are obtained using both methods. The distributional properties of these estimators are studied in large samples, via asymptotic theory, and in small and moderate samples, via Monte Carlo simulation. The distribution is shown to be appropriate for a wide variety of meteorological data.

Monitoring baseline suspended sediment in forested basins: the effects of sampling on suspended sediment rating curves

Abstract: Rating curves are widely used for directly assessing changes in the suspended sediment delivery process and indirectly for estimating total yields. Four sampling methods were simulated over a 31-day record of suspended sediment from the North Fork of the Mad River near Korbel, California. The position and size of the four groups of plotted slope/intercept pairs indicated differences in bias and variance among the methods. Estimates of total yield for the 31-day period and for storms of three sizes were also biased according to sampling method. A standard bias-correcting technique improved yield estimates, but did not remove sampling bias uniformly. Methods of data collection have a large and systematic effect on the estimation of rating-curve parameters and on estimates of suspended sediment yield. Differences attributed to land management may, in fact, result from changes in sampling methods.

Sampling Techniques for Gravel Sized Sediments

Abstract: The methods commonly used for sampling the coarser surface layers of gravel‐bed streams are reviewed. It is found that while an areal sample is biased toward the coarser grains compared with a volumetric sample of the same material, the conversion formula suggested by a voidless cube model overcompensates for this effect. A modified version of the cube model that accounts for porosity is proposed. The modified cube model and tests conducted with areal sampling using wax indicate that area‐by‐weight analyses can be converted successfully to the equivalent bulk sieve analyses. The average depth of the wax samples ranged from D65-D91, increasing with median grain size. The equivalence of the grid by number and bulk sieve analyses is reaffirmed.

Estimating soil loss on topographically non-uniform field and farm units

Abstract: Several forms of the universal soil loss equation, along with the ANSWERS hydrologic/erosion model were used to estimate soil erosion on topographically non-uniform field and farm units on two farms in Indiana. Both soil loss at a point and average soil loss along a landscape profile were used to estimate erosion. Soil loss estimates using USLE methods were aggregated to obtain field and farm averages. ANSWERS was included because it computes erosion over topographically non-uniform areas. Sampling density was critical for the USLE methods, while element size was critical for ANSWERS. With sufficiently dense sampling, each USLE method gave similar results, but the method that estimated average erosion on the landscape profiles required the fewest sampling points. The ANSWERS and USLE results, although different, were comparable considering the differences in prediction techniques.