The Experts below are selected from a list of 25272 Experts worldwide ranked by ideXlab platform
Andreas Savin - One of the best experts on this subject based on the ideXlab platform.
-
probabilistic performance estimators for computational chemistry methods the empirical Cumulative Distribution Function of absolute errors
Journal of Chemical Physics, 2018Co-Authors: Pascal Pernot, Andreas SavinAbstract:Benchmarking studies in computational chemistry use reference datasets to assess the accuracy of a method through error statistics. The commonly used error statistics, such as the mean signed and mean unsigned errors, do not inform end-users on the expected amplitude of prediction errors attached to these methods. We show that, the Distributions of model errors being neither normal nor zero-centered, these error statistics cannot be used to infer prediction error probabilities. To overcome this limitation, we advocate for the use of more informative statistics, based on the empirical Cumulative Distribution Function of unsigned errors, namely, (1) the probability for a new calculation to have an absolute error below a chosen threshold and (2) the maximal amplitude of errors one can expect with a chosen high confidence level. Those statistics are also shown to be well suited for benchmarking and ranking studies. Moreover, the standard error on all benchmarking statistics depends on the size of the reference dataset. Systematic publication of these standard errors would be very helpful to assess the statistical reliability of benchmarking conclusions.
-
probabilistic performance estimators for computational chemistry methods the empirical Cumulative Distribution Function of absolute errors
Journal of Chemical Physics, 2018Co-Authors: Pascal Pernot, Andreas SavinAbstract:Benchmarking studies in computational chemistry use reference datasets to assess the accuracy of a method through error statistics. The commonly used error statistics, such as the mean signed and mean unsigned errors, do not inform end-users on the expected amplitude of prediction errors attached to these methods. We show that, the Distributions of model errors being neither normal nor zero-centered, these error statistics cannot be used to infer prediction error probabilities. To overcome this limitation, we advocate for the use of more informative statistics, based on the empirical Cumulative Distribution Function of unsigned errors, namely, (1) the probability for a new calculation to have an absolute error below a chosen threshold and (2) the maximal amplitude of errors one can expect with a chosen high confidence level. Those statistics are also shown to be well suited for benchmarking and ranking studies. Moreover, the standard error on all benchmarking statistics depends on the size of the reference dataset. Systematic publication of these standard errors would be very helpful to assess the statistical reliability of benchmarking conclusions.Benchmarking studies in computational chemistry use reference datasets to assess the accuracy of a method through error statistics. The commonly used error statistics, such as the mean signed and mean unsigned errors, do not inform end-users on the expected amplitude of prediction errors attached to these methods. We show that, the Distributions of model errors being neither normal nor zero-centered, these error statistics cannot be used to infer prediction error probabilities. To overcome this limitation, we advocate for the use of more informative statistics, based on the empirical Cumulative Distribution Function of unsigned errors, namely, (1) the probability for a new calculation to have an absolute error below a chosen threshold and (2) the maximal amplitude of errors one can expect with a chosen high confidence level. Those statistics are also shown to be well suited for benchmarking and ranking studies. Moreover, the standard error on all benchmarking statistics depends on the size of the referenc...
Michael J Seiler - One of the best experts on this subject based on the ideXlab platform.
-
determinants of the strategic mortgage default Cumulative Distribution Function
Journal of Real Estate Literature, 2016Co-Authors: Michael J SeilerAbstract:In this paper, I discuss the many factors and considerations that enter into the strategic mortgage default (SMD) decision-making process. While it is not possible to construct a single Cumulative ...
-
determinants of the strategic mortgage default Cumulative Distribution Function
Social Science Research Network, 2014Co-Authors: Michael J SeilerAbstract:This study discusses the many factors and considerations that enter into the strategic mortgage default (SMD) decision-making process. While it is not possible to construct a single Cumulative Distribution Function (CDF) associated with this decision, it is important for policymakers to better understand what composes its shape as well as the range over which values occur. In this paper, I use both transactions-based and experiment-based data to suggest a theoretical shape of the SMD CDF.
Pascal Pernot - One of the best experts on this subject based on the ideXlab platform.
-
probabilistic performance estimators for computational chemistry methods the empirical Cumulative Distribution Function of absolute errors
Journal of Chemical Physics, 2018Co-Authors: Pascal Pernot, Andreas SavinAbstract:Benchmarking studies in computational chemistry use reference datasets to assess the accuracy of a method through error statistics. The commonly used error statistics, such as the mean signed and mean unsigned errors, do not inform end-users on the expected amplitude of prediction errors attached to these methods. We show that, the Distributions of model errors being neither normal nor zero-centered, these error statistics cannot be used to infer prediction error probabilities. To overcome this limitation, we advocate for the use of more informative statistics, based on the empirical Cumulative Distribution Function of unsigned errors, namely, (1) the probability for a new calculation to have an absolute error below a chosen threshold and (2) the maximal amplitude of errors one can expect with a chosen high confidence level. Those statistics are also shown to be well suited for benchmarking and ranking studies. Moreover, the standard error on all benchmarking statistics depends on the size of the reference dataset. Systematic publication of these standard errors would be very helpful to assess the statistical reliability of benchmarking conclusions.
-
probabilistic performance estimators for computational chemistry methods the empirical Cumulative Distribution Function of absolute errors
Journal of Chemical Physics, 2018Co-Authors: Pascal Pernot, Andreas SavinAbstract:Benchmarking studies in computational chemistry use reference datasets to assess the accuracy of a method through error statistics. The commonly used error statistics, such as the mean signed and mean unsigned errors, do not inform end-users on the expected amplitude of prediction errors attached to these methods. We show that, the Distributions of model errors being neither normal nor zero-centered, these error statistics cannot be used to infer prediction error probabilities. To overcome this limitation, we advocate for the use of more informative statistics, based on the empirical Cumulative Distribution Function of unsigned errors, namely, (1) the probability for a new calculation to have an absolute error below a chosen threshold and (2) the maximal amplitude of errors one can expect with a chosen high confidence level. Those statistics are also shown to be well suited for benchmarking and ranking studies. Moreover, the standard error on all benchmarking statistics depends on the size of the reference dataset. Systematic publication of these standard errors would be very helpful to assess the statistical reliability of benchmarking conclusions.Benchmarking studies in computational chemistry use reference datasets to assess the accuracy of a method through error statistics. The commonly used error statistics, such as the mean signed and mean unsigned errors, do not inform end-users on the expected amplitude of prediction errors attached to these methods. We show that, the Distributions of model errors being neither normal nor zero-centered, these error statistics cannot be used to infer prediction error probabilities. To overcome this limitation, we advocate for the use of more informative statistics, based on the empirical Cumulative Distribution Function of unsigned errors, namely, (1) the probability for a new calculation to have an absolute error below a chosen threshold and (2) the maximal amplitude of errors one can expect with a chosen high confidence level. Those statistics are also shown to be well suited for benchmarking and ranking studies. Moreover, the standard error on all benchmarking statistics depends on the size of the referenc...
Jose F Paris - One of the best experts on this subject based on the ideXlab platform.
-
on the bivariate nakagami m Cumulative Distribution Function closed form expression and applications
IEEE Transactions on Communications, 2013Co-Authors: F J Lopezmartinez, David Moralesjimenez, Eduardo Martosnaya, Jose F ParisAbstract:In this paper, we derive exact closed-form expressions for the bivariate Nakagami-m Cumulative Distribution Function (CDF) with positive integer fading severity index m in terms of a class of hypergeometric Functions. Particularly, we show that the bivariate Nakagami-m CDF can be expressed as a finite sum of elementary Functions and bivariate confluent hypergeometric Φ3 Functions. Direct applications which arise from the proposed closed-form expression are the outage probability (OP) analysis of a dual-branch selection combiner in correlated Nakagami-m fading, or the calculation of the level crossing rate (LCR) and average fade duration (AFD) of a sampled Nakagami-m fading envelope.
-
closed form expressions for rician shadowed Cumulative Distribution Function
Electronics Letters, 2010Co-Authors: Jose F ParisAbstract:New analytical results are presented for the Cumulative Distribution Function (CDF) of Rician shadowed random variables. In particular, these results find applicability in the performance analysis of land-mobile satellite (LMS) communications.
Yunlung Teng - One of the best experts on this subject based on the ideXlab platform.
-
adaptive sampling based on the Cumulative Distribution Function of order statistics to delineate heavy metal contaminated soils using kriging
Environmental Pollution, 2005Co-Authors: Kaiwei Juang, Daryuan Lee, Yunlung TengAbstract:Abstract Correctly classifying “contaminated” areas in soils, based on the threshold for a contaminated site, is important for determining effective clean-up actions. Pollutant mapping by means of kriging is increasingly being used for the delineation of contaminated soils. However, those areas where the kriged pollutant concentrations are close to the threshold have a high possibility for being misclassified. In order to reduce the misclassification due to the over- or under-estimation from kriging, an adaptive sampling using the Cumulative Distribution Function of order statistics (CDFOS) was developed to draw additional samples for delineating contaminated soils, while kriging. A heavy-metal contaminated site in Hsinchu, Taiwan was used to illustrate this approach. The results showed that compared with random sampling, adaptive sampling using CDFOS reduced the kriging estimation errors and misclassification rates, and thus would appear to be a better choice than random sampling, as additional sampling is required for delineating the “contaminated” areas.