Search NASA⌕ Search

SEARCH · Search NASA

Results for “sample size”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 1,009 records · Page 56

Extreme Mean and Its Applications

Extreme value statistics obtained from normally distributed data are considered. An extreme mean is defined as the mean of p-th probability truncated normal distribution. An unbiased estimate of this extreme mean and its large sample distribution are derived. The distribution of this estimate even for very large samples is found to be nonnormal. Further, as the sample size increases, the variance of the unbiased estimate converges to the Cramer-Rao lower bound. The computer program used to obtain the density and distribution functions of the standardized unbiased estimate, and the confidence intervals of the extreme mean for any data are included for ready application. An example is included to demonstrate the usefulness of extreme mean application.

Swaroop, R.↗

On evaluating clustering procedures for use in classification

The problem of evaluating clustering algorithms and their respective computer programs for use in a preprocessing step for classification is addressed. In clustering for classification the probability of correct classification is suggested as the ultimate measure of accuracy on training data. A means of implementing this criterion and a measure of cluster purity are discussed. Examples are given. A procedure for cluster labeling that is based on cluster purity and sample size is presented.

Pore, M. D.↗

Mass-Loss Buttons Monitor Material Degradation

Small button-sized samples attached to parent materials are simple way of monitoring degradation of parent in harsh environments. Samples determine effects of multiple exposures to environmental extremes without disturbing fit or function of parent. They are less costly and more convenient than complex instrumentation normally required to measure complete temperature/pressure time history of parent component.

Webster, C. N.↗

A contribution to the knowledge of HMX decomposition and application of results

The decomposition of cyclotramethylenetetranitramine (HMX) in the solid and liquid phase was studied by isothermal and nonisothermal heating at atmospheric pressure. Decomposition rates of solid HMX changed with sample size and gaseous environment. Kinetic parameters were obtained from weight loss measurements in the temperature range 229 C - 269 C. These tests also yielded highly porous solid residues. Qualitative aspects of solid and liquid phase decomposition of HMX with additives were also investigated in isothermal and nonisothermal tests.

Kraeutle, K. J.↗

Sympathetic reaction tests and analyses

The critical separation distances for explosive reactions of a solid rocket propellant were measured. Explosive reactions included low order explosion, low order detonation, and high order detonation. The effects of sample size, shape, damage and temperature on sympathetic reaction were determined experimentally. The sympathetic detonation of small cubes of solid rocket propellant was modelled numerically, using the Eulerian reactive hydrodynamic code 2DE with Forest Fire burn rates. The model was applied to cubes of 2.54 - 7.62 cm (1 - 3 in.), with agreement between calculated and experimental results.

Ricardson, D. E.↗

Artifacts in digital images

Three kinds of artifacts unique to digital images are illustrated, namely aliasing caused by undersampling, interference phenomena caused by improper display of images, and harmonic overtones caused by quantization of amplitudes. Special attention is given to undersampling when the sample size and interval are the same. It is noted that this situation is important because it is typical of solid-state cameras. Quantization of image data of necessity introduces energy at harmonic overtones of the image spectrum. This energy is aliased if the frequency of the overtones is greater than 0.5 cycle/pixel. It cannot be selectively removed from the image through filtering, and the best way to suppress it is to maximize the amplification of the sensor before digital encoding.

Lorre, J. J.↗

Freshwater ice thickness observations using passive microwave sensors

Walden Reservoir, a freshwater lake in north-central Colorado, was overflown six times by a NASA C-130 aircraft between January 1977 and April 1980. The aircraft was equipped with four microwave radiometers operating between 0.81 and 6.0 cm in wavelength (37.0 to 5.0 GHz). The 6.0-cm radiometer data showed a good relationship with ice thickness based on a sample of four ice thickness values. The 1.67- and 1.35-cm radiometer data showed weaker relationships with ice thickness. The 0.81-cm sensor data showed no positive relationship with ice thickness. None of the relationships was statistically significant because of the small sample size. The 6.0-cm sensor data in the nadir-viewing mode was found to have the most potential of all the wavelengths studied, for use in remotely determining ice thickness. The 6.0-cm radiometer probably sensed the entire thickness of the ice on the reservoir (ranging from 25.4 to 67.3 cm in thickness) and was apparently not significantly affected by the snow overlying the ice. The shorter wavelengths are scattered by the snow overlying the ice and are more suitable for snow studies than for ice thickness studies.

Hall, D. K.↗

Computer program documentation for the patch subsampling processor

The programs presented are intended to provide a way to extract a sample from a full-frame scene and summarize it in a useful way. The sample in each case was chosen to fill a 512-by-512 pixel (sample-by-line) image since this is the largest image that can be displayed on the Integrated Multivariant Data Analysis and Classification System. This sample size provides one megabyte of data for manipulation and storage and contains about 3% of the full-frame data. A patch image processor computes means for 256 32-by-32 pixel squares which constitute the 512-by-512 pixel image. Thus, 256 measurements are available for 8 vegetation indexes over a 100-mile square.

Nieves, M. J.↗

Radioactivities in returned lunar materials and in meteorites

Carbon-14 measurements were made for meteorites with a Van der Graaf accelerator. Accelerator C-14 dating improved the precision by a factor of ten, allowed the use of smaller sample sizes, and gave speedier results than C-14 dating with counters. A methodology for determining the terrestrial ages of several antarctic meteorites is described and the results are listed.

Fireman, E. L.↗

Software reliability: A comparison of results obtained from established software reliability models

Two models of the software error detection process are compared, the Jelinski-Moranda model and a Bayes inference model. Simulation techniques are used to generate software related system failure data which is analyzed by both models. Point estimates and confidence limits are compared. It is demonstrated that uncertainty may be considerable for reasonable samples sizes and should be considered in any application of these techniques. The Jelinski-Moranda model is sensitive to the failure of data to follow internal assumptions of the model, often not providing any point estimates, a factor which may limit its usefulness in many real world situations. The Bayes model is shown to respond to the introduction of additional errors in the software correction process, a condition where error counting models such as the Jelinski-Moranda generally fail to converge.

Horn, M. H.↗

Software models: A Bayesian approach to parameter estimation in the Jelenski-Moranda software reliability model

Maximum likelihood estimation procedures for the Jelinski-Moranda software reliability model often give misleading answers. A reparameterization and a Bayesian analysis eliminate some of the problems incurred by MLE methods and often give better predictions on sets of real and simulated data. Practical difficulties in estimating the initial number of errors N and the failure rate of each error phi by the method of maximum likelihood are: N, the MLE of N, is occasionally infinite (i.e., the routines for calculating N and phi do not converge). It is shown that N is finite sub i only if the regression line of the interevent times t sub i vs. i has positive slope. A serious problem is that often N approximates n, the sample size, and sometimes N = n. Thus the MLE predicts that the program is perfect even when it is far from being so. Only when almost all failures have been removed can N and phi be trusted near the end of debugging.

Littlewood, B.↗

Pyrolysis of Precambrian kerogens - Constraints and capabilities

Precambrian kerogens are currently considered to be the primary candidates for the search of biochemical fossils. Degradation of kerogens by relatively 'mild' pyrolysis techniques, such as under high vacuum, can liberate indicative structural moieties which were incorporated in, and perhaps shielded by, these solid and highly condensed, basically aromatic substances. It is necessary to observe analytical constraints (sample size and shape, temperature, pressure, time, etc.) in order to prevent an overabundant yield of secondary pyrolyzates (inter- and intramolecular rearrangements) which can prevent kerogen characterization. Potential biochemical fossils have been found in Precambrian kerogens. Demonstratable syngenetic biochemical fossils are expected after kerogen diagenesis and catagenesis is understood in sufficient detail, and when pyrolysis is augmented by multiple, improved analytical techniques.

Nagy, B.↗

Sampling errors in the estimation of empirical orthogonal functions

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.

North, G. R.↗

The Comstar D/3 gain degradation experiment

The results of gain degradation measurements using the Comstar D/3 19.04 GHz beacon are reported. This experiment utilized 0.6 and 5 m aperture antennas aligned along the same propagation path to examine propagation effects which are related to the antenna aperture size. Sample data for clear air, scintillation in clear air, and precipitation fading are presented. Distributions of the received signal levels and variances for both antennas are also presented.

Lee, T. C.↗

Development of test methodology for dynamic mechanical analysis instrumentation

Dynamic mechanical analysis instrumentation was used for the development of specific test methodology in the determination of engineering parameters of selected materials, esp. plastics and elastomers, over a broad range of temperature with selected environment. The methodology for routine procedures was established with specific attention given to sample geometry, sample size, and mounting techniques. The basic software of the duPont 1090 thermal analyzer was used for data reduction which simplify the theoretical interpretation. Clamps were developed which allowed 'relative' damping during the cure cycle to be measured for the fiber-glass supported resin. The correlation of fracture energy 'toughness' (or impact strength) with the low temperature (glassy) relaxation responses for a 'rubber-modified' epoxy system was negative in result because the low-temperature dispersion mode (-80 C) of the modifier coincided with that of the epoxy matrix, making quantitative comparison unrealistic.

Allen, V. R.↗

A Poisson process approximation for generalized K-5 confidence regions

One-sided confidence regions for continuous cumulative distribution functions are constructed using empirical cumulative distribution functions and the generalized Kolmogorov-Smirnov distance. The band width of such regions becomes narrower in the right or left tail of the distribution. To avoid tedious computation of confidence levels and critical values, an approximation based on the Poisson process is introduced. This aproximation provides a conservative confidence region; moreover, the approximation error decreases monotonically to 0 as sample size increases. Critical values necessary for implementation are given. Applications are made to the areas of risk analysis, investment modeling, reliability assessment, and analysis of fault tolerant systems.

Arsham, H.↗

Accuracy of remotely sensed data: Sampling and analysis procedures

A review and update of the discrete multivariate analysis techniques used for accuracy assessment is given. A listing of the computer program written to implement these techniques is given. New work on evaluating accuracy assessment using Monte Carlo simulation with different sampling schemes is given. The results of matrices from the mapping effort of the San Juan National Forest is given. A method for estimating the sample size requirements for implementing the accuracy assessment procedures is given. A proposed method for determining the reliability of change detection between two maps of the same area produced at different times is given.

Congalton, R. G.↗

Current collection from the space plasma through defects in high voltage solar array insulation

For spacecraft operation in the near Earth environment, solar cell arrays constitute the major source of reliable long term power. Optimization of mass and power efficiency results in a general requirement for high voltage solar arrays. The space plasma environment, though, can result in large currents being collected by exposed solar cells. The solution of a protective covering of transparent insulation is not a complete solution, inasmuch as defects in the insulation result in anomalously large currents being collected through the defects. Tests simulating the electron collection from small defects in an insulation have shown that there are two major collection modes. The first mode involves current enhancement by means of a surface phenomenon involving the surrounding insulator. In the second mode the current collection is enhanced by vaporization and ionization of the insulators materials, in addition to the surface enhancement of the first mode. A model for the electron collection is the surface enhanced collection mode was developed. The model relates the secondary electron emission yield to the electron collection. It correctly predicts the qualitative effects of hole size, sample temperature and roughening of sample surface. The theory was also shown to predict electron collection within a factor of two for the polymers teflon and polyimide.

Stillwell, R. P.↗