Search NASA⌕ Search

SEARCH · Search NASA

Results for “Bayesian Testing”

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 19 records

Bayesian Hypothesis Testing

These are a set of slides about Bayesian hypothesis testing, where many hypotheses are tested. The conclusions are the following: The value of the Bayes factor obtained when using the median of the posterior marginal is almost the minimum value of the Bayes factor. The value of τ 2 which minimizes the Bayes factor is a reasonable choice for this parameter. This allows a likelihood ratio to be computed with is the least favorable to H 0 .

97 MATHEMATICS AND COMPUTING↗

Hierarchical Model for Bayesian Hypothesis Testing

A common problem in data analysis is testing a point hypothesis. For example, one might test whether or not the mean of a distribution is zero. We describe the standard Bayesian approach, which requires the choice of a width parameter to define a prior distribution for an alternative hypothesis. Standard practice is to choose this width parameter to be the least favorable choice for the original, point hypothesis. This approach, however, is ad-hoc and not informed by the data. We examine this practice in more detail than in previous work by constructing a hierarchical model, where the width parameter is an unknown value subject to inference. This yields a marginal posterior distribution for the width parameter. We then show that using the median of this marginal posterior distribution as the width parameter in the Bayesian comparison between the hypotheses gives results which are very similar to those obtained from the existing approaches. This gives a more rigorous and data-informed justification for the standard practice of choosing a width parameter which is least favorable to the point hypothesis.

97 MATHEMATICS AND COMPUTING↗

Using a Simple Binomial Model to Assess Improvement in Predictive Capability: Sequential Bayesian Inference, Hypothesis Testing, and Power Analysis

We present a Bayesian statistical methodology for identifying improvement in predictive simulations, including an analysis of the number of (presumably expensive) simulations that will need to be made in order to establish with a given level of confidence that an improvement has been observed. Our analysis assumes the ability to predict (or postdict) the same experiments with legacy and new simulation codes and uses a simple binomial model for the probability, {theta}, that, in an experiment chosen at random, the new code will provide a better prediction than the old. This model makes it possible to do statistical analysis with an absolute minimum of assumptions about the statistics of the quantities involved, at the price of discarding some potentially important information in the data. In particular, the analysis depends only on whether or not the new code predicts better than the old in any given experiment, and not on the magnitude of the improvement. We show how the posterior distribution for {theta} may be used, in a kind of Bayesian hypothesis testing, both to decide if an improvement has been observed and to quantify our confidence in that decision. We quantify the predictive probability that should be assigned, prior to taking any data, to the possibility of achieving a given level of confidence, as a function of sample size. We show how this predictive probability depends on the true value of {theta} and, in particular, how there will always be a region around {theta} = 1/2 where it is highly improbable that we will be able to identify an improvement in predictive capability, although the width of this region will shrink to zero as the sample size goes to infinity. We show how the posterior standard deviation may be used, as a kind of 'plan B metric' in the case that the analysis shows that {theta} is close to 1/2 and argue that such a plan B should generally be part of hypothesis testing. All the analysis presented in the paper is done with a general beta-function prior for {theta}, enabling sequential analysis in which a small number of new simulations may be done and the resulting posterior for {theta} used as a prior to inform the next stage of power analysis.

42 ENGINEERING↗

Impact of genotype‐calling methodologies on genome‐wide association and genomic prediction in polyploids

Abstract Discovery and analysis of genetic variants underlying agriculturally important traits are key to molecular breeding of crops. Reduced representation approaches have provided cost‐efficient genotyping using next‐generation sequencing. However, accurate genotype calling from next‐generation sequencing data is challenging, particularly in polyploid species due to their genome complexity. Recently developed Bayesian statistical methods implemented in available software packages, polyRAD, EBG, and updog, incorporate error rates and population parameters to accurately estimate allelic dosage across any ploidy. We used empirical and simulated data to evaluate the three Bayesian algorithms and demonstrated their impact on the power of genome‐wide association study (GWAS) analysis and the accuracy of genomic prediction. We further incorporated uncertainty in allelic dosage estimation by testing continuous genotype calls and comparing their performance to discrete genotypes in GWAS and genomic prediction. We tested the genotype‐calling methods using data from two autotetraploid species, Miscanthus sacchariflorus and Vaccinium corymbosum , and performed GWAS and genomic prediction. In the empirical study, the tested Bayesian genotype‐calling algorithms differed in their downstream effects on GWAS and genomic prediction, with some showing advantages over others. Through subsequent simulation studies, we observed that at low read depth, polyRAD was advantageous in its effect on GWAS power and limit of false positives. Additionally, we found that continuous genotypes increased the accuracy of genomic prediction, by reducing genotyping error, particularly at low sequencing depth. Our results indicate that by using the Bayesian algorithm implemented in polyRAD and continuous genotypes, we can accurately and cost‐efficiently implement GWAS and genomic prediction in polyploid crops.

59 BASIC BIOLOGICAL SCIENCES↗

Bayesian Design of Experiments for Logistic Regression to Evaluate Multiple Forensic Algorithms

When evaluating the performance of several forensic classification algorithms, it is desirable to construct a design that considers a variety of performance levels for each of the algorithms. We describe a strategy to use Bayesian design of experiments with multiple prior estimates to capture anticipated performance. Our goal is to characterize results from the different classification algorithms as a function of multiple explanatory variables and use this to choose a design about which units to test. Bayesian design of experiments has been successful for generalized linear models, including logistic regression models. Here, we develop methodology for the case where there are several potentially nonoverlapping priors for anticipated performance under consideration. The weighted priors method performs well for a broad range of true underlying model parameter choices and is more robust when compared to other candidate design choices. Additionally, we show how this can be applied in the multivariate input case and provide some useful summary measures. The shared information plot is used to evaluate design point allocation, and the D-value difference plot allows for the comparison of design performance across multiple potential parameter values in higher dimensions. Here, we illustrate the methods with several examples.

97 MATHEMATICS AND COMPUTING↗

The weighted priors approach for combining expert opinions in logistic regression experiments

When modeling the reliability of a system or component, it is not uncommon for more than one expert to provide very different prior estimates of the expected reliability as a function of an explanatory variable such as age or temperature. Our goal in this paper is to incorporate all information from the experts when choosing a design about which units to test. Bayesian design of experiments has been shown to be very successful for generalized linear models, including logistic regression models. We use this approach to develop methodology for the case where there are several potentially non-overlapping priors under consideration. While multiple priors have been used for analysis in the past, they have never been used in a design context. The Weighted Priors method performs well for a broad range of true underlying model parameter choices and is more robust when compared to other reasonable design choices. Finally, we illustrate the method through multiple scenarios and a motivating example. Additional figures for this article are available in the online supplementary information.

97 MATHEMATICS AND COMPUTING↗

On the detection of spectral ripples from the recombination epoch

Photons emitted during cosmological hydrogen (500≲z≲1600) and helium recombination (1600≲z≲3500 for He ii → He i, 5000≲z≲8000 for He iii → He ii) are predicted to appear as broad, weak spectral distortions of the cosmic microwave background. We present a feasibility study for a ground-based detection of these recombination lines, which would uniquely probe astrophysical cosmology beyond the last scattering surface and provide observational constraints on the thermal history of the universe. We find that including sufficient signal spectral structure and maximizing signal-to-noise ratio, an octave band in the 2–6 GHz window is optimal; in this band the predicted signal appears as an additive quasi-sinusoidal component with amplitude about 8 nK embedded in a sky spectrum some nine orders of magnitude brighter. We discuss algorithms to detect these tiny spectral fluctuations in the sky spectrum by foreground modeling and introduce a maximally smooth function capable of describing the foreground spectrum and distinguishing the signal of interest. We conclude that detection is in principle feasible in realistic observing times provided that radio frequency interference and instrument bandpass calibration are controlled in this band at the required level; using Bayesian tests and mock data, we show that 90% confidence detection is possible with an array of 128 radiometers observing for 255 days of effective integration time. We propose APSERa—Array of Precision Spectrometers for the Epoch of Recombination—a dedicated radio telescope to detect these recombination lines.

79 ASTRONOMY AND ASTROPHYSICS↗

The NANOGrav nine-year data set: Limits on the isotropic stochastic gravitational wave background

We compute upper limits on the nanohertz-frequency isotropic stochastic gravitational wave background (GWB) using the 9 year data set from the North American Nanohertz Observatory for Gravitational Waves (NANOGrav) collaboration. Well-tested Bayesian techniques are used to set upper limits on the dimensionless strain amplitude (at a frequency of 1 yr{sup −1}) for a GWB from supermassive black hole binaries of A{sub gw}<1.5×10{sup −15}. We also parameterize the GWB spectrum with a broken power-law model by placing priors on the strain amplitude derived from simulations of Sesana and McWilliams et al. Using Bayesian model selection we find that the data favor a broken power law to a pure power law with odds ratios of 2.2 and 22 to one for the Sesana and McWilliams prior models, respectively. Using the broken power-law analysis we construct posterior distributions on environmental factors that drive the binary to the GW-driven regime including the stellar mass density for stellar-scattering, mass accretion rate for circumbinary disk interaction, and orbital eccentricity for eccentric binaries, marking the first time that the shape of the GWB spectrum has been used to make astrophysical inferences. Returning to a power-law model, we place stringent limits on the energy density of relic GWs, Ω{sub gw}(f)h{sup 2}<4.2 × 10{sup −10}. Our limit on the cosmic string GWB, Ω{sub gw}(f)h{sup 2}<2.2 × 10{sup −10}, translates to a conservative limit on the cosmic string tension with Gμ<3.3×10{sup −8}, a factor of four better than the joint Planck and high-l cosmic microwave background data from other experiments.

79 ASTRONOMY AND ASTROPHYSICS↗

The NANOGrav Nine-Year Data Set: Limits on the Isotropic Stochastic Gravitational Wave Background

We compute upper limits on the nanohertz-frequency isotropic stochastic gravitational wave background (GWB) using the 9 year data set from the North American Nanohertz Observatory for Gravitational Waves (NANOGrav) collaboration. Well-tested Bayesian techniques are used to set upper limits on the dimensionless strain amplitude (at a frequency of 1 yr(exp -1) for a GWB from supermassive black hole binaries of A(sub gw) less than 1.5 x 10(exp -15). We also parameterize the GWB spectrum with a broken power-law model by placing priors on the strain amplitude derived from simulations of Sesana and McWilliams et al. Using Bayesian model selection we find that the data favor a broken power law to a pure power law with odds ratios of 2.2 and 22 to one for the Sesana and McWilliams prior models, respectively. Using the broken power-law analysis we construct posterior distributions on environmental factors that drive the binary to the GW-driven regime including the stellar mass density for stellar-scattering, mass accretion rate for circumbinary disk interaction, and orbital eccentricity for eccentric binaries, marking the first time that the shape of the GWB spectrum has been used to make astrophysical inferences. Returning to a power-law model, we place stringent limits on the energy density of relic GWs, OMEGA(sub gw) (f) h squared less than 4.2 x 10(exp -10). Our limit on the cosmic string GWB, OMEGA(sub gw) (f) h squared less than 2.2 x 10(exp -10), translates to a conservative limit on the cosmic string tension with G mu less than 3.3 x 10(exp -8), a factor of four better than the joint Planck and high-l‚ cosmic microwave background data from other experiments.

Arzoumanian, Z.↗

Testing T2K’s Bayesian constraints with priors in alternate parameterisations

Bayesian analysis results require a choice of prior distribution. In long-baseline neutrino oscillation physics, the usual parameterisation of the mixing matrix induces a prior that privileges certain neutrino mass and flavour state symmetries. Here we study the effect of privileging alternate symmetries on the results of the T2K experiment. We find that constraints on the level of CP violation (as given by the Jarlskog invariant) are robust under the choices of prior considered in the analysis. On the other hand, the degree of octant preference for the atmospheric angle depends on which symmetry has been privileged.

Bayesian Inference↗

Automated Monitoring with a BCP Fault-Decision Test

The Bayesian conditional probability (BCP) technique is a statistical fault-decision technique that is suitable as the mathematical basis of the fault-manager module in the automated-monitoring system and method described in the immediately preceding article. Within the automated-monitoring system, the fault-manager module operates in conjunction with the fault-detector module, which can be based on any one of several fault-detection techniques; examples include a threshold-limit-comparison technique or the BSP or SPRT technique mentioned in the preceding article. The present BCP technique is used to evaluate a series of one or more fault-detection events for the purpose of filtering out occasional false alarms produced by many types of statistical fault-detection procedures. The BCP technique increases the probability that an automated monitoring system produces a correct decision regarding the presence or absence of a fault. Because occasional false alarms are an inevitable consequence of the SPRT, BSP, or any other statistically based fault-detection test, there is a need for a logical procedure to distinguish between true and false alarms. Heretofore, it has been common practice to make a fault decision on an ad hoc basis for example by following a multiple-observation voting strategy in which a signal is declared to be indicative of a fault if m of the last n observations produced a fault-detection alarm. The BCP technique was developed to obtain results more reliable than those afforded by a voting strategy. The BCP technique involves a test in which one applies Bayesian inference techniques to a series of one or more single-observation alarms produced by a fault-detection test. One considers the last n decisions generated by a fault-detection test in order to evaluate the conditional probability that a failure is indicated (see figure). Each new decision reached by a fault-detection test is treated as a new piece of evidence about the state of the monitored asset, and the conditional probability of failure for the system is updated on the basis of this new evidence. The conditional probability of failure is compared with a predefined limit. For a probability below the limit, the asset is declared to be healthy. For a probability above the limit, the asset is declared to be faulty.

Bickford, Randall L.↗

Finding inputs that trigger floating-point exceptions in heterogeneous computing via Bayesian optimization

Testing code for floating-point exceptions is crucial as exceptions can quickly propagate and produce unreliable numerical answers. The state-of-the-art to test for floating-point exceptions in heterogeneous systems is quite limited and solutions require the application’s source code, which precludes their use in accelerated libraries where the source is not publicly available. We present an approach to find inputs that trigger floating-point exceptions in black-box CPU or GPU functions, i.e., functions where the source code and information about input bounds are unavailable. Our approach is the first to use Bayesian optimization (BO) to identify such inputs and uses novel strategies to overcome the challenges that arise in applying BO to this problem. Here, we implement our approach in the XSCOPE framework and demonstrate it on 58 functions from the CUDA Math Library and 81 functions from the Intel Math Library. XSCOPE is able to identify inputs that trigger exceptions in about 73% of the tested functions.

97 MATHEMATICS AND COMPUTING↗

Testing general relativity using Bayesian model selection: Applications to observations of gravitational waves from compact binary systems

Second-generation interferometric gravitational-wave detectors, such as Advanced LIGO and Advanced Virgo, are expected to begin operation by 2015. Such instruments plan to reach sensitivities that will offer the unique possibility to test general relativity in the dynamical, strong-field regime and investigate departures from its predictions, in particular, using the signal from coalescing binary systems. We introduce a statistical framework based on Bayesian model selection in which the Bayes factor between two competing hypotheses measures which theory is favored by the data. Probability density functions of the model parameters are then used to quantify the inference on individual parameters. We also develop a method to combine the information coming from multiple independent observations of gravitational waves, and show how much stronger inference could be. As an introduction and illustration of this framework-and a practical numerical implementation through the Monte Carlo integration technique of nested sampling-we apply it to gravitational waves from the inspiral phase of coalescing binary systems as predicted by general relativity and a very simple alternative theory in which the graviton has a nonzero mass. This method can (and should) be extended to more realistic and physically motivated theories.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Bayes and discovery

Objective Bayesian Hypothesis Testing approach is presented.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Statistical Significance Testing for Mixed Priors: A Combined Bayesian and Frequentist Analysis

In many hypothesis testing applications, we have mixed priors, with well-motivated informative priors for some parameters but not for others. The Bayesian methodology uses the Bayes factor and is helpful for the informative priors, as it incorporates Occam’s razor via the multiplicity or trials factor in the look-elsewhere effect. However, if the prior is not known completely, the frequentist hypothesis test via the false-positive rate is a better approach, as it is less sensitive to the prior choice. We argue that when only partial prior information is available, it is best to combine the two methodologies by using the Bayes factor as a test statistic in the frequentist analysis. We show that the standard frequentist maximum likelihood-ratio test statistic corresponds to the Bayes factor with a non-informative Jeffrey’s prior. We also show that mixed priors increase the statistical power in frequentist analyses over the maximum likelihood test statistic. We develop an analytic formalism that does not require expensive simulations and generalize Wilks’ theorem beyond its usual regime of validity. In specific limits, the formalism reproduces existing expressions, such as the p-value of linear models and periodograms. We apply the formalism to an example of exoplanet transits, where multiplicity can be more than 10 7 . We show that our analytic expressions reproduce the $p$-values derived from numerical simulations. We offer an interpretation of our formalism based on the statistical mechanics. We introduce the counting of states in a continuous parameter space using the uncertainty volume as the quantum of the state. We show that both the $p$-value and Bayes factor can be expressed as an energy versus entropy competition.

97 MATHEMATICS AND COMPUTING↗

Designing and testing inflationary models with Bayesian networks

Even simple inflationary scenarios have many free parameters. Beyond the variables appearing in the inflationary action, these include dynamical initial conditions, the number of fields, and couplings to other sectors. These quantities are often ignored but cosmological observables can depend on the unknown parameters. We use Bayesian networks to account for a large set of inflationary parameters, deriving generative models for the primordial spectra that are conditioned on a hierarchical set of prior probabilities describing the initial conditions, reheating physics, and other free parameters. We use N{sub f}-quadratic inflation as an illustrative example, finding that the number of e-folds N{sub *} between horizon exit for the pivot scale and the end of inflation is typically the most important parameter, even when the number of fields, their masses and initial conditions are unknown, along with possible conditional dependencies between these parameters.

79 ASTRONOMY AND ASTROPHYSICS↗