Showing papers on "Sampling (statistics) published in 1982"
TL;DR: In this paper, the authors reviewed the necessary weighting factors for gridded data and the sampling errors incurred when too small a sample is available, and a rule of thumb indicating when an EOF is likely to be subject to large sampling fluctuations is presented.
Abstract: Empirical Orthogonal Functions (EOF's), eigenvectors of the spatial cross-covariance matrix of a meteorological field, are reviewed with special attention given to the necessary weighting factors for gridded data and the sampling errors incurred when too small a sample is available. The geographical shape of an EOF shows large intersample variability when its associated eigenvalue is “close” to a neighboring one. A rule of thumb indicating when an EOF is likely to be subject to large sampling fluctuations is presented. An explicit example, based on the statistics of the 500 mb geopotential height field, displays large intersample variability in the EOF's for sample sizes of a few hundred independent realizations, a size seldom exceeded by meteorological data sets.
2,793 citations
Book•
01 Jan 1982
TL;DR: In this article, the one-way classification mutiple comparisons analysis of variance is used for the analysis of covariance in a normal distribution comparison involving two sample means, where the covariance is defined as the probability of the two samples having the same distribution.
Abstract: Observations probability sampling from a normal distribution comparisons involving two sample means principles of experimental design analysis of variance I - the one-way classification mutiple comparisons analysis of variance II - multiway classification linear regression linear correlation matrix notation linear regression in matrix notation multiple and partial regression and correlation analysis of variance III - factorial experiments analysis of variance analysis of covariance IV analysis of covariance analysis of variance V - unequal subclass numbers some uses of chi-square enumeration data I - one-way classifications enumeration data II - contingency tables categorical models some discrete distributions nonparametric statistics sampling finite populations.
693 citations
TL;DR: In this paper, a 2-dimensional (2-D) Gabor expansion on a set of basic wavelets adapted to phase preservation is proposed to handle instantaneous frequency spectra.
Abstract: Morlet et al (1982, this issue) showed the advantages of using complex values for both waves and characteristics of the media. We simulated the theoretical tools we present here, using the Goupillaud-Kunetz algorithm. Now we present sampling methods for complex signals or traces corresponding to received waves, and sampling methods for complex characterization of multilayered or heterogeneous media. Regarding the complex signals, we present a twodimecsional(2-D) method of sampling in the time-frequency domain using a special or “extended” Gabor expansion on a set of basic wavelets adapted to phase preservation. Such a 2-D expansion permits us to handle in a proper manner instantaneous frequency spectra. We show the differences between “wavelet resolution” and “sampling grid resolution.” We also show the importance of phase preservation in high-resolution seismic. Regarding the media, we show how analytical studies of wave propagation in periodic structured layers could help when trying to characterize the physical properties of the layers and their large scale granularity as a result of complex deconvolution. Analytical studies of wave propagation in periodic structures are well known in solid state physics, and lead to the so-called “Bloch waves.” The introduction of complex waves leads to replacing the classical wave equation by a Schriidinger equation. Finally, we show that complex wave equations, Gabor expansion, and Bloch waves are three different ways of ‘introducing the tools of quantum mechanics in highresolution seismic (Gabor, 1946; Kittel, 1976, Morlet, 1975). And conversely, the Goupillaud-Kunetz algorithm and an extended Gabor expansion may be of some use in solid state physics.
595 citations
Journal Article•
TL;DR: In this article, it is shown how one can use knowledge of map-category relative sizes to improve estimates of various probabilities, by means of two simple sampling plans suggested in the accuracy assessment literature.
Abstract: By means of two simple sampling plans suggested in the accuracy-assessment literature, it is shown how one can use knowledge of map-category relative sizes to improve estimates of various probabilities. The fact that maximum likelihood estimates of cell probabilities for the simple random sampling and map category-stratified sampling were identical has permitted a unified treatment of the contingency-table analysis. A rigorous analysis of the effect of sampling independently within map categories is made possible by results for the stratified case. It is noted that such matters as optimal sample size selection for the achievement of a desired level of precision in various estimators are irrelevant, since the estimators derived are valid irrespective of how sample sizes are chosen.
359 citations
TL;DR: In this paper, a stochastic method for global optimization is described and evaluated, which involves a combination of sampling, clustering and local search, with a range of confidence intervals on the value of the global optimum.
Abstract: A stochastic method for global optimization is described and evaluated. The method involves a combination of sampling, clustering and local search, and terminates with a range of confidence intervals on the value of the global optimum. Computational results on standard test functions are included as well.
263 citations
TL;DR: In this article, the maximum likelihood estimator is derived and shown to be readily calculated using an iterative procedure that starts with the Mayfield (1975) estimate as a trial value.
Abstract: Statistical methods for estimating and comparing constant survival rates are developed here for sampling designs in which survival of a subject is checked at irregular intervals. The maximum likelihood estimator is derived and shown to be readily calculated using an iterative procedure that starts with the Mayfield (1975) estimate as a trial value. Sampling distributions of this estimator and of the product of two or more estimates are skewed, and normalizing transformations are provided to facilitate valid confidence interval estimation. The sampling distribution of the difference between two independent estimates is found to be sufficiently normal without transformation to allow valid use of conventional normal theory procedures for testing differences and determining sample size for specified power. Statistical validity under the variable intensity sampling design does require that the duration of intervisit periods vary independently of observer perceptions concerning the survival status of the subject and, in order to achieve robustness with respect to the assumption of constant survivorship, sampling intensity must vary independently of any temporal changes in the daily survival rate. Investigators are warned no, to return earlier than planned to subjects thought to have died, as this observer behavior may cause serious bias in the survivorship estimate.
232 citations
01 Jan 1982
229 citations
210 citations
TL;DR: In this paper, Glimm's method is extended to several space variables by operator splitting, using van der Corput sampling sequence to obtain high resolution results both in smooth parts of the solution, as well as at discontinuities.
Abstract: We investigate Glimm's method, a method for constructing approximate solutions to systems of hyperbolic conservation laws in one space variable by sampling explicit wave solutions It is extended to several space variables by operator splitting We consider two problems 1) We propose a highly accurate form of the sampling procedure, in one space variable, based on the van der Corput sampling sequence We test the improved sampling procedure numerically in the case of inviscid compressible flow in one space dimension and find that it gives high resolution results both in the smooth parts of the solution, as well as at discontinuities 2) We investigate the operator splitting procedure by means of which the multidimensional method is constructed An $O(1)$ error stemming from the use of this procedure near shocks oblique to the spatial grid is analyzed numerically in the case of the equations for inviscid compressible flow in two space dimensions We present a hybrid method which eliminates this error, consisting of Glimm's method, used in continuous parts of the flow, and the nonlinear Godunov method, used in regions where large pressure jumps are generated The resulting method is seen to be a substantial improvement over either of the component methods for multidimensional calculations
209 citations
Book•
01 Dec 1982
TL;DR: In this paper, the Horvitz-Thompson Estimator is used to estimate the probability of an individual sample having an equal probability of being selected for sampling with unequal probabilities without replacement.
Abstract: 1: An Introduction to Sampling with Unequal Probabilities.- 2: Descriptions of Procedures for Sampling with Unequal Probabilities without Replacement.- 3: Unequal Probability Sampling Procedures and the Horvitz-Thompson Estimator.- 4: Selection Procedures Using Special Estimators.- 5: Multistage Sampline.- 5: An Optimal Sampling Strategy for Large Unistage Samples.- Appendices.- Appendix C: Overlap Between Poisson and Collocated Samples.- Selection Procedures' Index.- Authors' Index.
193 citations
TL;DR: Suggested refinements of the USFWS standard procedure to improve efficiency include use of an inexpensive plaster disc saturated with attractant, permitting rapid volatilization, and computer analysis of data by a program incorporating the Fisher Randomization Test and Wilcoxon Signed Rank Test.
Abstract: The scent-station index method used by the U.S. Fish and Wildlife Service (USFWS) and others is a practical means for determining trends in carnivore populations. Suggested refinements of the USFWS standard procedure to improve efficiency include (1) use of an inexpensive plaster disc saturated with attractant, permitting rapid volatilization; (2) lines of 10 scent stations operated for 1 night to increase sample sizes, enhance sampling distribution, and minimize weather interference; and (3) computer analysis of data by a program incorporating the Fisher Randomization Test and Wilcoxon Signed Rank Test, which avoid unwarranted assumptions and yield greater sensitivity to changes in visitation rates than that afforded by the Z test used previously. The Fisher Randomization Test is a particularly powerful and efficient tool. The computer program, applicable to any paired data, is described. Scent-station survey recommendations based on these refinements and our experience are given. J. WILDL. MANAGE. 46(1):217-229 Interest in carnivore, especially coyote (Canis latrans), ecology and management dictates a need for assessing carnivore abundance. Because direct enumeration of the animals is impractical, the U.S. Fish and Wildlife Service, in its West-wide survey of predator abundance, assesses trends on the basis of animal visitation to lines of artificial scent stations. Each station is a circle of sifted earth having, in the center, a perforated plastic capsule filled with an odor attractant. Tracks visible in the sifted earth are recorded and smoothed over daily by an observer (Linhart and Knowlton 1975, Roughton and Sweeny 1979). Efficiency and reliability of the scentstation method are influenced by the quality and quantity of the odor attractant used, method of presentation, sampling design, and sensitivity of the data analysis. Attractant quality and the selection of fatty acid scent (FAS) are discussed in a companion paper (Roughton 1982). The presen paper reports upon tests of (1) volatilization rates of FAS from several substrates, (2) several methods of presenting FAS at scent stations, (3) relative s atistical and time efficiencies of several sampling designs (i.e., combinations of number of stations per line 'and number of nights the lines are in operation), and (4) alternative methods of analysis for inferring differences in visitation rates. The final section summarizes our recommendations from these studies and 8 years of experience with the scent-station method. The first 2 objectives were directed at increasing scent-station visitation, to reduce the estimate of variance in comparing results from different sampling periods (Hodges 1975). Theoretically, the optimal range for detecting percent change in visitation rates is 40-60% (D. Anderson and C. Romesburg, pers. commun.), well above the West-wide average of 10% for the USFWS survey. Conjecture about the influence of the amount of attractant released into the atmosphere led to studies of volatilization rates, even though we do not understand the inter1 Present address: Brigantine National Wildlife Refuge, U.S. Fish and Wildlife Service, Oceanville, NJ 08231. J. Wildl. Manage. 46(1):1982 217 This content downloaded from 157.55.39.159 on Sun, 18 Sep 2016 06:27:00 UTC All use subject to http://about.jstor.org/terms 218 SCENT-STATION METHODOLOGYRoughton and Sweeny action between quality (in terms of odor components) and quantity (in terms of molecules/m3 of air vs. thresholds of detection). Originally, a plastic capsule served to protect the powdered attractant and limited the amount used at each station. These capsules are expensive, accounting for 65% of the material costs (Roughton 1980), and are inconvenient to use. We sought a better and less costly device for presenting attractant. The 3rd objective related to optimizing the sampling effort within manpower and road-network constraints. The current design (Roughton and Sweeny 1979) employs lines of 50 scent stations observed for 4 days. Because stations are only 0.5 km apart and coyotes occasionally visit consecutive stations, stations cannot be regarded as independent sampling units. Likewise, field observation and scrutiny of visitation patterns in our survey data suggested individual carnivores may visit the same (or nearby) stations on more than 1 night, so nights are also not independent sampling units. For analytical purposes, the interrelationship of scentstation visits spatially and among nights delimits the sampling unit as "each line of 50 stations run for 4 days" (a 50 x 4 line). Because each observer checks, at most, 2 lines in a week, the maximum data yield is 2 sampling units/man-week. Increased sampling efficiency seemed desirable. The 4th objective was to determine the most powerful appropriate statistical test for scent-station data. The only statistical analyses routinely performed for our survey are statewide and West-wide comparisons of coyote visitation rates for succeeding years. Coyote trend analyses for areas within states are legitimate, but would not yield useful results because of inadequate sample sizes. Even at the statewide level, statistically significant changes in coyote visitation are rarely detected between years, even though the percent changes are sometimes striking. We believe this results from inadequate sample sizes under the current survey design and an inherent lack of power in the Z test applied to the USFWS survey data. The Z test (Roughton and Sweeny 1979) assumes that: the survey lines are independent, which they are; the data are normally distributed, which is uncertain; and the variance for a given line is homogeneous from year to year, which is likely only under stable population conditions. The rarity of stable conditions is the root of the problem, because the estimated variance for a given line is based on the data for all years in which that line was run, even though part of the year-toyear variability may result from true changes in population density, the parameter we are trying to estimate. We believe this test lacks satisfactory power to detect the year-to-year changes (which are of primary interest), requires dubious assumptions, and relies upon overly conservative estimates of variance. We thank C. E. Harris, J. T. Roughton, B. G. Wagner, Jr., and L. D. Walker for assistance in data collection; D. Alexander, C. Becenti, M. Kaschke, E. Olson, C. D. Sykes, M. Walker, J. Wilbrecht, and L. A. Windberg for cooperation and advice in selecting study areas; D. C. Bowden, J. I. Hodges, K. Marshall, and C. Romesburg for advice and assistance in statistical analysis and computer programming; and F. F. Knowlton and L. C. Stoddart for criticism of the manuscript.
TL;DR: This article examines and empirically compares the components of total survey error for several research designs and data collection methods and consistent finding is that nonsampling error is the major contributor tototal survey error, while random sampling error is minimal.
Abstract: To increase the validity and reliability of survey data, one must minimize total error and its components, sampling and nonsampling error. This article examines and empirically compares the compone...
TL;DR: Growth rate estimates calculated from the maximum mitotic index in phased populations are shown to be accurate when the duration of the mitotic stage is greater than the interval during which the increase in cell number occurs.
Abstract: A review of existing approaches used to calculate phytoplankton population growth rates from fractions of cells undergoing mitosis (mitotic index) has revealed inconsistencies in their content and underlying assumptions. The appropriate expressions are derived for populations in exponential growth and those which exhibit phased cell division. Growth rate estimates calculated from the maximum mitotic index in phased populations are shown to be accurate when the duration of the mitotic stage is greater than the interval during which the increase in cell number occurs. This characteristic is manifested in a plateau-shaped curve of mitotic index vs. time. The alternative approach of averaging sequential observations is also considered, and the conditions for which one can obtain exact results are defined. The magnitude of uncertainties is examined as a function of the frequency of sampling.
TL;DR: In this article, a multifacet sampling model based on generalizability theory is developed for the measurement of dispositional attributes, where dispositions are defined in terms of universes of observations, and the value of the disposition is given by the universe score, the mean over the universe defining the disposition.
Abstract: A multifacet sampling model, based on generalizability theory, is developed for the measurement of dispositional attributes. Dispositions are defined in terms of universes of observations, and the value of the disposition is given by the universe score, the mean over the universe defining the disposition. Observed scores provide estimates of universe scores, and errors of measurement are introduced in order to maintain consistency in these estimates. The sampling model provides a straightforward interpretation of validity in terms of the accuracy of estimates of the universe scores, and of reliability in terms of the consistency among these estimates. A third property of measurements, import, is defined in terms of all of the implications of a measurement. The model provides the basis for a detailed analysis of standardization and of the systematic errors that standardization creates; for example, the hypothesis that increases in reliability may cause decreases in validity is easily derived from the model...
TL;DR: In this paper, the authors define microdebitage as all stone flaking residue less than 1 mm in maximum dimension and demonstrate that it is produced in great quantities by stone tool manufacture and can permeate site matrices as a permanent signature of past cultural activity.
TL;DR: The extent of axillary lymph-node involvement was assessed in 50 consecutive cases of operable breast cancer by means ofAxillary sampling as well as complete axillary clearance in every case and the sampling method has an error rate of 24% and is therefore unreliable in predicting nodal involvement.
TL;DR: The method can be used when the time interval between samples and when VS vary from sample to sample, but return of the drug from the perfused body to the perfusate after sampling may cause C'i to be overestimated.
TL;DR: A hardware implementation is described which achieves 70 dB spurious-free dynamic range and a bandwidth of 30 kHz and the bandpass to low-pass translation is accomplished by undersampling.
Abstract: Many radio and sonar systems require signal outputs in complex low-pass form. To achieve this, it is possible to use uniform sampling of the bandpass signal, together with computation of the quadrature component by way of a Hilbert transform. The bandpass to low-pass translation is accomplished by undersampling. A hardware implementation is described which achieves 70 dB spurious-free dynamic range and a bandwidth of 30 kHz.
TL;DR: This paper is a 1-D analysis of the degradation caused by image sampling and interpolative reconstruction that includes the sample-scene phase as an explicit random parameter and provides a complete characterization of this image degradation as the sum of two terms.
Abstract: This paper is a 1-D analysis of the degradation caused by image sampling and interpolative reconstruction. The analysis includes the sample-scene phase as an explicit random parameter and provides a complete characterization of this image degradation as the sum of two terms: one term accounts for the mean effect of undersampling (aliasing) and nonideal reconstruction averaged over all sample-scene phases; the other term accounts for variations about this mean. The results of this paper have application to the design and performance analysis of image scanning, sampling, and reconstruction systems.
TL;DR: An opportunity for exposure--disease-based sampling arises whenever data on exposure and disease are both available on a group of subjects who are incompletely characterized in terms of other covariates.
Abstract: When study subjects are obtained through sampling procedures based on the level of known characteristics, covariate data can be used to adjust effect estimates. This principle, which is well-understood for both exposure-based sampling (cohort studies) and disease-based sampling (case-control studies), can also be applied to cross-classifications of the source population. An opportunity for exposure--disease-based sampling arises whenever data on exposure and disease are both available on a group of subjects who are incompletely characterized in terms of other covariates. This is a common intermediate stage in epidemiologic studies of many kinds.
01 Feb 1982
TL;DR: Independently generated input data was used to verify that interfailure times are very nearly exponentially distributed and to obtain good estimates of the failure rates of individual errors and demonstrate how widely they vary.
Abstract: A software experiment conducted with repetitive run sampling is reported Independently generated input data was used to verify that interfailure times are very nearly exponentially distributed and to obtain good estimates of the failure rates of individual errors and demonstrate how widely they vary This fact invalidates many of the popular software reliability models now in use The log failure rate of interfailure time was nearly linear as a function of the number of errors corrected A new model of software reliability is proposed that incorporates these observations
TL;DR: Timing comparisons with the best available algorithms for sampling the gamma distribution show the new algorithms to be acceptably fast.
Abstract: Computer algorithms are described for simulation of the generalized inverse Gaussian, generalized hyperbolic and hyperbolic distributions The efficiencies of the algorithms are found Timing comparisons with the best available algorithms for sampling the gamma distribution show the new algorithms to be acceptably fast The extension to sampling multivariate generalized hyperbolic distributions is escribed Listings of Fortran implementations of the algorithms are available
TL;DR: In this paper, the authors examine sampling issues that should be better addressed in conducting family studies and discuss the use of probability and non-probability techniques for drawing samples their advantages and disadvantages and examples of the types of samples that can be drawn from public records on issues involving marriage and the family.
Abstract: This paper examines sampling issues that should be better addressed in conducting family studies. The focus is on systematic sampling biases that cannot be handled as adequately using statistical techniques as can random sources of error. The topics include the use of probability and nonprobability techniques for drawing samples their advantages and disadvantages and examples of the types of samples that can be drawn from public records on issues involving marriage and the family. Sampling biases produced by inadequate coverage of the sample selected are also discussed. Common problems in sampling are illustrated through a review of selected U.S. studies on divorce. Suggestions are also offered regarding ways to produce a more standard description of sampling issues. (EXCERPT)
TL;DR: In this paper, the existence of horizontal as well as vertical gradients, and their importance in water quality sampling design were the subject of intensive transect sampling efforts at DeGray Lake, a U.S. Army Corps of Engineers reservoir in southern Arkansas.
Abstract: The design of monitoring programs often serves as one of the major sources of error or uncertainty in water quality data. Properly designed programs should minimize uncertainty or at least provide a means by which variability can be partitioned into recognizable components. While the design of sampling programs has received recent attention, commonly employed strategies for limnological sampling of lakes may not be completely appropriate for many reservoirs. Based on NES data, reservoirs are generally larger, deeper, and morphologically more complex than natural lakes. Reservoirs also receive a majority of their inflow from a single tributary located a considerable distance from the point of outflow. The result is the establishment of marked physical, biological, and chemical gradients from headwater to dam. The existence of horizontal as well as vertical gradients, and their importance in water quality sampling design were the subject of intensive transect sampling efforts at DeGray Lake, a U.S. Army Corps of Engineers reservoir in southern Arkansas. Data collected were used to partition Variance, identify areas of similarity, and demonstrate how an equitable sampling program might be designed.
TL;DR: In this article, a method for estimating the sampling variance of estimates of mean daily energy expenditure of free-ranging animals, based on time-energy budgets, was proposed, which combines estimated components of variation from laboratory and field observations into a single estimate of the overall sample variance.
Abstract: In this paper I outline a method for estimating the sampling variance of estimates of mean daily energy expenditure of free-ranging animals, based on time-energy budgets. This method combines estimated components of variation from laboratory and field observations into a single estimate of the overall sample variance. An approximate confidence region for the mean daily ex- penditure can be constructed as well. I illustrate these methods by applying them to the energy budget of the Golden-winged Sunbird (Nectarinia reichenowi) originally published by Gill and Wolf (1975). The total sample variance of the mean daily energy expenditure, when that expenditure is esti- mated from the time-energy budget approach, is a sum of many terms, including the products of the variances of individual variables, the squared estimates of the mean of individual variables, and the products of the estimated means and covariances among the variables. Thus, the estimated sample variance will usually be large. Any biases in estimation will be propagated throughout the calculations, and errors of staggering magnitude can result. Knowledge of the components of total sampling vari- ance and the sensitivity of these estimates to the various sources of bias should aid field biologists in designing time-energy budget studies so that the maximum possible reliability can be achieved.