Estimation in truncated Poisson distributions with concomitant exposure intervals and truncated points Final report
Estimation in truncated Poisson distributions with concomitant exposure intervals and truncation points
Cohen, A. C.↗
SEARCH · Search NASA
Results for “Truncated distribution”
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.
Electron propagation in the leaky box model with a truncated pathlength distribution
A study of electron propagation using the leaky box model is discussed. It is shown that a truncated pathlength distribution due to a lack of nearby sources is responsible for the steepening of the electron spectrum. The electron spectrum is broken into three regions: a low-energy region where electron storage is dominated by escape, a medium energy region in which the electron energy loss lifetime is sufficiently short to dominate propagation, but not so short as to prevent electrons from propagating throughout the storage region, and a third energy region in which the energy loss lifetime is shorter than the time it takes for cosmic rays to diffuse to earth from the nearest source. The asymptotic slope of the electron spectrum is shown to be steeper than that of the injection spectrum by more than one power of energy.
Mauger, B. G.↗
Use of the negative binomial-truncated Poisson distribution in thunderstorm prediction
A probability model is presented for the distribution of thunderstorms over a small area given that thunderstorm events (1 or more thunderstorms) are occurring over a larger area. The model incorporates the negative binomial and truncated Poisson distributions. Probability tables for Cape Kennedy for spring, summer, and fall months and seasons are presented. The computer program used to compute these probabilities is appended.
Cohen, A. C.↗
Application of a truncated normal failure distribution in reliability testing
Statistical truncated normal distribution function is applied as a time-to-failure distribution function in equipment reliability estimations. Age-dependent characteristics of the truncated function provide a basis for formulating a system of high-reliability testing that effectively merges statistical, engineering, and cost considerations.
Groves, C., Jr.↗
Improving Land Cover Product-Based Estimates of the Extent of Fragmented Cover Types
The effects of changing land use/land cover on regional and global climate ecosystems depends on accurate estimates of the extent of critical land cover types such as Arctic wetlands and fire scars in boreal forests. To address this information requirement, land cover products at coarse spatial resolution such as Advanced Very High Resolution Radiometer (AVHRR) -based maps and the MODIS Land Cover Product are being produced. The accuracy of the extent of highly fragmented cover types such as fire scars and ponds is in doubt because much (the numerous scars and ponds smaller than the pixel size) is missed. A promising method for improving areal estimates involves modeling the observed distribution of the fragment sizes as a type of truncated distribution, then estimating the sum of unobserved sizes in the lower, truncated tail and adding it to the sum of observed fragment sizes. The method has been tested with both simulated and actual cover products.
Hlavka, Christine A.↗
Effect of bow-type initial imperfection on reliability of minimum-weight, stiffened structural panels
Computations were performed to determine the effect of an overall bow-type imperfection on the reliability of structural panels under combined compression and shear loadings. A panel's reliability is the probability that it will perform the intended function - in this case, carry a given load without buckling or exceeding in-plane strain allowables. For a panel loaded in compression, a small initial bow can cause large bending stresses that reduce both the buckling load and the load at which strain allowables are exceeded; hence, the bow reduces the reliability of the panel. In this report, analytical studies on two stiffened panels quantified that effect. The bow is in the shape of a half-sine wave along the length of the panel. The size e of the bow at panel midlength is taken to be the single random variable. Several probability density distributions for e are examined to determine the sensitivity of the reliability to details of the bow statistics. In addition, the effects of quality control are explored with truncated distributions.
Stroud, W. Jefferson↗
An advanced probabilistic structural analysis method for implicit performance functions
In probabilistic structural analysis, the performance or response functions usually are implicitly defined and must be solved by numerical analysis methods such as finite element methods. In such cases, the most commonly used probabilistic analysis tool is the mean-based, second-moment method which provides only the first two statistical moments. This paper presents a generalized advanced mean value (AMV) method which is capable of establishing the distributions to provide additional information for reliability design. The method requires slightly more computations than the second-moment method but is highly efficient relative to the other alternative methods. In particular, the examples show that the AMV method can be used to solve problems involving non-monotonic functions that result in truncated distributions.
Wu, Y.-T.↗
Numerical solution of the Navier-Stokes equations for laminar, transonic flows
An implicit finite difference solution of the Navier-Stokes equations yielded time histories of the transonic laminar flow development about a circular cylinder and NACA-0018 airfoil. Reynolds numbers ranged from those corresponding to purely laminar flow to those corresponding to significant turbulence in the boundary layer. Body thermal conditions of an adiabatic wall and a specified body temperature were considered. Versatility in treating arbitrary bodies was incorporated by using numerically generated, body-fitted coordinate transformations. Solution of the simultaneous difference equations for the dependent variables was obtained using an accelerated Gauss-Seidel iterative scheme. Computational results are presented in the form of velocity vector fields, Mach number contours, aerodynamic coefficients, heat transfer rates at the body surface, and body temperature distributions. Truncation analyses of first and second derivative difference approximations resulted in general criteria for numerically generated coordinates so that flow near a body is more accurately represented.
Turner, L., III↗
Reliability of stiffened structural panels: Two examples
The reliability of two graphite-epoxy stiffened panels that contain uncertainties is examined. For one panel, the effect of an overall bow-type initial imperfection is studied. The size of the bow is assumed to be a random variable. The failure mode is buckling. The benefits of quality control are explored by using truncated distributions. For the other panel, the effect of uncertainties in a strain-based failure criterion is studied. The allowable strains are assumed to be random variables. A geometrically nonlinear analysis is used to calculate a detailed strain distribution near an elliptical access hole in a wing panel that was tested to failure. Calculated strains are used to predict failure. Results are compared with the experimental failure load of the panel.
Stroud, W. Jefferson↗
Pseudorandom noise for telemetry error rate measurement applications and limitations.
A pseudorandom noise (PRN) generator functional design has been developed which is superior to the commercially available instrument in usable bandwidth and in the closeness of fit to the gaussian probability distribution in the 'tails' of the curve. The principal disadvantage of PRN sources for bit error rate (BER) measurements is that the probability distribution is truncated in the tails. This truncation of the probability distribution results in a similar truncation of the BER curve - i.e., there is a minimum BER which can be measured, below which the only possible result is zero. The minimum nonzero BER actually computed with the PRN source developed was 0.00001 with a pseudorandom sequence of 131,071 bits. It is theoretically possible, with a longer sequence, to achieve a minimum BER of 0.000001 with the same design.
Hopkins, P. M.↗
On the Distribution of the Sum of Independent Doubly Truncated Gamma Variables
Pearson distribution of doubly truncated gamma variables
Lavender, D. E.↗
Truncated Gaussians as tolerance sets
This work focuses on the use of truncated Gaussian distributions as models for bounded data measurements that are constrained to appear between fixed limits. The authors prove that the truncated Gaussian can be viewed as a maximum entropy distribution for truncated bounded data, when mean and covariance are given. The characteristic function for the truncated Gaussian is presented; from this, algorithms are derived for calculation of mean, variance, summation, application of Bayes rule and filtering with truncated Gaussians. As an example of the power of their methods, a derivation of the disparity constraint (used in computer vision) from their models is described. The authors' approach complements results in Statistics, but their proposal is not only to use the truncated Gaussian as a model for selected data; they propose to model measurements as fundamentally in terms of truncated Gaussians.
Cozman, Fabio↗
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.↗
A simple test of independence for truncated data with applications to redshift surveys
This paper presents an easily applied permutation test for H0, closely related to Lyden-Bell's (1971) estimate of the marginal distribution of truncated data. The test is applied to two redshift-magnitude surveys, one of galaxies and one of quasars. Assuming statistical independence, testing H0 amounts to testing validity of the cosmological model. Segal's (1986) chronomatic cosmological model is rejected under H0. On the other hand, for the quasar sample H0 is rejected strongly in a conventional cosmological model (and in a chronomatic model as well) indicating either incorrectness of the models or, as is more commonly assumed, indicating strong luminosity evolution.
Efron, Bradley↗
Tables of maximum likelihood estimating functions for singly truncated and singly censored samples from the normal distribution
Tables of maximum likelihood estimating functions for singly truncated and singly censored samples from normal distribution
Cohen, A. C.↗
Rotational behavior of the main-sequence stars and its plausible consequences concerning formation of planetary system. II.
Statistical models based on truncated Maxwell distributions of original stellar angular momenta, discussing main sequence star rotational behavior and planetary system formation
Huang, S.-S.↗
Optimum structural design based on reliability and proof-load testing
Proof-load test eliminates structures with strength less than the proof load and improves the reliability value in analysis. It truncates the distribution function of strength at the proof load, thereby alleviating verification of a fitted distribution function at the lower tail portion where data are usually nonexistent.
Shinozuka, M.↗
Quasi-exospheric heat flux of solar-wind electrons
Density, bulk-velocity, and heat-flow moments are calculated for truncated Maxwellian distributions representing the cool and hot populations of solar-wind electrons, as realized at the base of a hypothetical exosphere. The electrostatic potential is thus calculated by requiring charge quasi-neutrality and the absence of electrical current. Plasma-kinetic coupling of the cool-electron and proton bulk velocities leads to an increase in the electrostatic potential and a decrease in the heat-flow moment.
Eviatar, A.↗