Search NASA⌕ Search

SEARCH · Search NASA

Results for “Chi-Square 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.

At least 37 records · Page 2
All sourcesNASA NTRSDOE OSTI
NASA NTRS1984

The distribution of hot spots

Statistical analyses are compared for two published hot spot data sets, one minimal set of 42 and another larger set of 117, using three different approaches. First, the earths surface is divided into 16 equal-area fractions and the observed distribution of hot spots among them is analyzed using chi-square tests. Second, cumulative distributions about the principal axes of the hot spot inertia tensor are used to describe hot spot distribution. Finally, a hot spot density function is constructed for each of the two hot spot data sets. The methods all indicate that hot spots have a nonuniform distribution, even when statistical fluctuations are considered. To the first order, hot spots are concentrated on one half of of the earth's surface area; within that portion, the distribution is consistent with a uniform distribution. The observed hot spot densities for neither data set are explained solely by plate speed.

Stefanick, M.↗
NASA NTRS1995

Solar cell angle of incidence corrections

Literature on solar array angle of incidence corrections was found to be sparse and contained no tabular data for support. This lack along with recent data on 27 GaAs/Ge 4 cm by 4 cm cells initiated the analysis presented in this paper. The literature cites seven possible contributors to angle of incidence effects: cosine, optical front surface, edge, shadowing, UV degradation, particulate soiling, and background color. Only the first three are covered in this paper due to lack of sufficient data. The cosine correction is commonly used but is not sufficient when the incident angle is large. Fresnel reflection calculations require knowledge of the index of refraction of the coverglass front surface. The absolute index of refraction for the coverglass front surface was not known nor was it measured due to lack of funds. However, a value for the index of refraction was obtained by examining how the prediction errors varied with different assumed indices and selecting the best fit to the set of measured values. Corrections using front surface Fresnel reflection along with the cosine correction give very good predictive results when compared to measured data, except there is a definite trend away from predicted values at the larger incident angles. This trend could be related to edge effects and is illustrated by a use of a box plot of the errors and by plotting the deviation of the mean against incidence angle. The trend is for larger deviations at larger incidence angles and there may be a fourth order effect involved in the trend. A chi-squared test was used to determine if the measurement errors were normally distributed. At 10 degrees the chi-squared test failed, probably due to the very small numbers involved or a bias from the measurement procedure. All other angles showed a good fit to the normal distribution with increasing goodness-of-fit as the angles increased which reinforces the very small numbers hypothesis. The contributed data only went to 65 degrees from normal which prevented any firm conclusions about extreme angle effects although a trend in the right direction was seen. Measurement errors were estimated and found to be consistent with the conclusions that were drawn. A controlled experiment using coverglasses and cells from the same lots and extending to larger incidence angles would probably lead to further insight into the subject area.

Burger, Dale R.↗
NASA NTRS1990

Estimation of mean rain rate - Application to satellite observations

A method for the estimation of the mean area average rain rate from dependent data is developed and applied to the GARP Atlantic Tropical Experiment data. The method consists of fitting a mixed distribution, containing an atom at zero, by minimum chi-square in combination with certain time-space sampling designs. In modeling the continuous component of the mixed distribution, it is shown that the lognormal distribution provides a very close fit for the nonzero area average rainrates. A comparison with the gamma distribution shows that the lognormal distribution is a better choice as expressed by the minimum chi-square criterion. Some of the time-space sampling designs correspond to satellite sampling. The results indicate that a satellite visiting an area of about 350 x 350 sq km in the tropics approximately every 10 hours over a period can provide a rather close estimate for the mean area average rain rate.

Kedem, Benjamin↗
NASA NTRS1993

Improving the chi-squared approximation for bivariate normal tolerance regions

Let X be a two-dimensional random variable distributed according to N2(mu,Sigma) and let bar-X and S be the respective sample mean and covariance matrix calculated from N observations of X. Given a containment probability beta and a level of confidence gamma, we seek a number c, depending only on N, beta, and gamma such that the ellipsoid R = (x: (x - bar-X)'S(exp -1) (x - bar-X) less than or = c) is a tolerance region of content beta and level gamma; i.e., R has probability gamma of containing at least 100 beta percent of the distribution of X. Various approximations for c exist in the literature, but one of the simplest to compute -- a multiple of the ratio of certain chi-squared percentage points -- is badly biased for small N. For the bivariate normal case, most of the bias can be removed by simple adjustment using a factor A which depends on beta and gamma. This paper provides values of A for various beta and gamma so that the simple approximation for c can be made viable for any reasonable sample size. The methodology provides an illustrative example of how a combination of Monte-Carlo simulation and simple regression modelling can be used to improve an existing approximation.

Feiveson, Alan H.↗
NASA NTRS1978

Comparison of high energy gamma rays from absolute value of b greater than 30 deg with the galactic neutral hydrogen distribution

High-energy gamma-ray (energy above 35 MeV) data from the SAS 2 satellite have been used to compare the intensity distribution of gamma rays with that of neutral hydrogen (H I) density along the line of sight, at high galactic latitudes (absolute values greater than 30 deg). A model has been constructed for the case where the observed gamma-ray intensity has been assumed to be the sum of a galactic component proportional to the H I distribution plus an isotropic extragalactic emission. A chi-squared test of the model parameters indicates that about 30% of the total high-latitude emission may originate within the Galaxy.

Ozel, M. E.↗
NASA NTRS1982

Updating Landsat-derived land-cover maps using change detection and masking techniques

The California Integrated Remote Sensing System's San Bernardino County Project was devised to study the utilization of a data base at a number of jurisdictional levels. The present paper discusses the implementation of change-detection and masking techniques in the updating of Landsat-derived land-cover maps. A baseline landcover classification was first created from a 1976 image, then the adjusted 1976 image was compared with a 1979 scene by the techniques of (1) multidate image classification, (2) difference image-distribution tails thresholding, (3) difference image classification, and (4) multi-dimensional chi-square analysis of a difference image. The union of the results of methods 1, 3 and 4 was used to create a mask of possible change areas between 1976 and 1979, which served to limit analysis of the update image and reduce comparison errors in unchanged areas. The techniques of spatial smoothing of change-detection products, and of combining results of difference change-detection algorithms are also shown to improve Landsat change-detection accuracies.

Likens, W.↗
NASA NTRS1992

Detecting change as it occurs

Traditionally climate changes have been detected from long series of observations and long after they have happened. Our 'inverse sequential' procedure, for detecting change as soon as it occurs, describes the existing or most recent data by their frequency distribution. Its parameter(s) are estimated both from the existing set of observations and from the same set augmented by 1,2,....j new observations. Individual-value probability products ('likelihoods') are used to form ratios which yield two probabilities for erroneously accepting the existing parameter(s) as valid for the augmented data set, and vice versa. A genuine parameter change is signaled when these probabilities (or a more stable compound probability) show a progressive decrease. New parameter values can then be estimated from the new observations alone using standard statistical techniques. The inverse sequential procedure will be illustrated for global annual mean temperatures (assumed normally distributed), and for annual numbers of North Atlantic hurricanes (assumed to represent Poisson distributions). The procedure was developed, but not yet tested, for linear or exponential trends, and for chi-squared means or degrees of freedom, a special measure of autocorrelation.

Radok, Uwe↗
NASA NTRS1988

Linear clusters of galaxies - A194

New measurements for 160 redshifts and previous measurements for 108 other redshifts are presented for galaxies within 5 deg of A194. The galaxy distribution in A194 is shown to be inconsistent with a spherically symmetric King model. A mass-to-light ratio is derived using the virial theorem which is lower than the mean for the groups in the CfA redshift survey (Huchra and Geller, 1982; Geller, 1984). A nonparametric test for galaxy-cluster alignment and a Chi-squared test are used to search for alignment of galaxy major axes with the axis of A194. Evidence for neither luminosity segregation nor significant differences in the velocity or surface distributions of galaxies as a function of morphological type is found.

Chapman, G. N. F.↗
NASA NTRS1987

Statistical tests for closure of plate motion circuits

Two methods, one based on a chi-square test and the second on an F-ratio test, of testing for the plate motion circuit closures are described and evaluated. The chi-square test is used to evaluate goodness of fit, and it is assumed that the assigned errors are accurate estimates of the true errors in the data. The F-ratio test is used to compare variances of distributions, and it is assumed that the relative values of assigned error are accurate. The two methods are applied to the data of Minster and Jordan (1978) on the motion of the three plates that meet at the Galapagos Triple Junction, and the motion of the three plates that meet at the Indian Ocean Triple Junction. It is noted that the F-ratio plate circuit closure test is more useful than the chi-square test for identifying systematic misfits in data because the chi-square test overestimates the errors of plate motion data.

Gordon, Richard G.↗
NASA NTRS1975

The effects of correlation on goodness of fit

Autocorrelated normal random variates were generated via computer and the effects of various levels of correlation on goodness of fit problems were studied. The results are useful in determining the distribution or estimating the parameters of populations that correlate observations such as wind speeds and temperature. The model used to generate the autocorrelated data is an autoregressive process of order 1. The Kolmogorov-Smirnov and chi-square statistics are used in the analysis. It was observed in the simulation that high positive correlations tend to shift the sample mean away from the population mean and negative correlations tend to shift the sample mean towards the population mean. In many cases, it was observed that positive and negative correlations tend to decrease the standard deviation. However, since this did not occur in all cases, no definite conclusion can be made regarding the standard deviation. Since the autoregressive process is a linear transformation, it is not surprising that normality was preserved. However, a possible extension of this problem could be to generate non-normal data and observe how the distribution is affected by correlation. Another extension might utilize another model such as autoregressive of order k or a moving average process of order k.

Kitchens, L. J.↗
NASA NTRS1987

Statistical analysis of secondary particle distributions in relativistic nucleus-nucleus collisions

The use is described of several statistical techniques to characterize structure in the angular distributions of secondary particles from nucleus-nucleus collisions in the energy range 24 to 61 GeV/nucleon. The objective of this work was to determine whether there are correlations between emitted particle intensity and angle that may be used to support the existence of the quark gluon plasma. The techniques include chi-square null hypothesis tests, the method of discrete Fourier transform analysis, and fluctuation analysis. We have also used the method of composite unit vectors to test for azimuthal asymmetry in a data set of 63 JACEE-3 events. Each method is presented in a manner that provides the reader with some practical detail regarding its application. Of those events with relatively high statistics, Fe approaches 0 at 55 GeV/nucleon was found to possess an azimuthal distribution with a highly non-random structure. No evidence of non-statistical fluctuations was found in the pseudo-rapidity distributions of the events studied. It is seen that the most effective application of these methods relies upon the availability of many events or single events that possess very high multiplicities.

Mcguire, Stephen C.↗
NASA NTRS1976

Multivariate normality

Sets of experimentally determined or routinely observed data provide information about the past, present and, hopefully, future sets of similarly produced data. An infinite set of statistical models exists which may be used to describe the data sets. The normal distribution is one model. If it serves at all, it serves well. If a data set, or a transformation of the set, representative of a larger population can be described by the normal distribution, then valid statistical inferences can be drawn. There are several tests which may be applied to a data set to determine whether the univariate normal model adequately describes the set. The chi-square test based on Pearson's work in the late nineteenth and early twentieth centuries is often used. Like all tests, it has some weaknesses which are discussed in elementary texts. Extension of the chi-square test to the multivariate normal model is provided. Tables and graphs permit easier application of the test in the higher dimensions. Several examples, using recorded data, illustrate the procedures. Tests of maximum absolute differences, mean sum of squares of residuals, runs and changes of sign are included in these tests. Dimensions one through five with selected sample sizes 11 to 101 are used to illustrate the statistical tests developed.

Crutcher, H. L.↗
NASA NTRS1973

Undergraduate research report

Studies are reported of the probability distribution of yield strength for mild steel, and the application of numerical techniques to solve mechanical vibration problems. Tensile experiments were conducted for 50 cold rolled and 38 hot rolled mild steel rods. Chi-square, Kolmogro-Smirnov and goodness-of-fit methods are used to test the normality or lognormality of the experimental results. As an example of the application of numerical techniques, a nonlinear dash-pot-speing system subjected to seismic excitation is analyzed by using the fourth-order Runge-Kutta integrated method.

Yeh, H. Y.↗
NASA NTRS1995

Gaussian statistics of the cosmic microwave background: Correlation of temperature extrema in the COBE DMR two-year sky maps

We use the two-point correlation function of the extrema points (peaks and valleys) in the Cosmic Background Explorer (COBE) Differential Microwave Radiometers (DMR) 2 year sky maps as a test for non-Gaussian temperature distribution in the cosmic microwave background anisotropy. A maximum-likelihood analysis compares the DMR data to n = 1 toy models whose random-phase spherical harmonic components a(sub lm) are drawn from either Gaussian, chi-square, or log-normal parent populations. The likelihood of the 53 GHz (A+B)/2 data is greatest for the exact Gaussian model. There is less than 10% chance that the non-Gaussian models tested describe the DMR data, limited primarily by type II errors in the statistical inference. The extrema correlation function is a stronger test for this class of non-Gaussian models than topological statistics such as the genus.

Kogut, A.↗
NASA NTRS1975

Tables of critical values for examining compositional non-randomness in proteins and nucleic acids

A binomially distributed statistic is defined to show whether or not the proportion of a particular amino acid in a protein deviates from random expectation. An analogous statistic is derived for nucleotides in nucleic acids. These new statistics are simply related to the classical chi-squared test. They explicitly account for the compositional fluctuations imposed by the finite length of proteins, and they are more accurate than previous tables.

Laird, M.↗
NASA NTRS1979

Soft X-ray structure of the Coma cluster of galaxies

Results are reported for an observation of the Coma cluster with an imaging X-ray telescope system aboard a sounding rocket. A two-dimensional X-ray map of the cluster in the energy range from 0.15 to 2.0 keV is presented. The radial distribution, granularity, and ellipticity of the X-ray surface brightness are investigated, and the spectrum and flux within the field of view are analyzed. These properties are then examined quantitatively by comparing the observations with models through the use of a chi-squared test. It is found that: (1) no single galaxy contributes more than a few percent of the total X-ray flux of the Coma cluster; (2) the X-ray source is diffuse but granular on a scale of several arcmin, particularly in the central region; (3) the centroid of the X-ray source is at a local intensity minimum; and (4) the data are most consistent with an isothermal hydrostatic model.

Gorentstein, P.↗
NASA NTRS1992

The polar cap structure of the X-ray pulsar 4U 1538 - 52

We compare the pulse-phase-dependent spectral observations of 4U 1538 - 52 obtained from Ginga to a series of static model atmospheres of curved accreting polar caps, utilizing a magnetized radiative transfer code with relativistic polarized cross sections. A chi-squared fit was performed on the symmetrized pulse shapes, allowing the parameters of the two polar caps to vary independently. We find evidence for a significant difference between the caps, both in terms of opening angle and temperature, while the optical depth is similar and compatible with radiative deceleration models. The polar caps appear larger than what is usually estimated from dipole fields, and are not located at the antipodes of each other. A broader distribution of magnetic fields is required than expected from a dipolar field at constant radius. If the magnetic field is indeed dipolar, it should be off-center in respect to the star and the accretion shock height exhibits an unusual dependence on the polar angle. An alternative possibility is that the magnetic field is disordered on scales small compared to the polar cap radius.

Bulik, T.↗