Publication detail

Global Interval Sensitivity Analysis of Hermite Probability Density Function Percentiles

KALA, Z.

Original Title

Global Interval Sensitivity Analysis of Hermite Probability Density Function Percentiles

English Title

Global Interval Sensitivity Analysis of Hermite Probability Density Function Percentiles

Type

journal article - other

Language

en

Original Abstract

The derivation of a Hermite probability density function whose parameters are mean value, standard deviation, skewness and kurtosis is presented in the article. Global interval sensitivity analysis capable of evaluating the effects of all combinations of the intervals of input parameters on the model output is presented. Global interval sensitivity analysis was used to identify the effects of uncertainties of the parameters of Hermite probability density function on its 5 percentile. A strong influence of the interaction between skewness and kurtosis, which would not have been identified using local sensitivity analysis, was identified. The global interval sensitivity analysis and Hermite probability density function presented here provide in combination with Monte Carlo simulation methods a very powerful tool for the probabilistic evaluation of structural reliability.

English abstract

The derivation of a Hermite probability density function whose parameters are mean value, standard deviation, skewness and kurtosis is presented in the article. Global interval sensitivity analysis capable of evaluating the effects of all combinations of the intervals of input parameters on the model output is presented. Global interval sensitivity analysis was used to identify the effects of uncertainties of the parameters of Hermite probability density function on its 5 percentile. A strong influence of the interaction between skewness and kurtosis, which would not have been identified using local sensitivity analysis, was identified. The global interval sensitivity analysis and Hermite probability density function presented here provide in combination with Monte Carlo simulation methods a very powerful tool for the probabilistic evaluation of structural reliability.

Keywords

Interval, interaction, kurtosis, skewnes, quantile, probability, reliability, sensitivity analysis

Released

15.10.2016

ISBN

1998-0140

Periodical

INTERNATIONAL JOURNAL of MATHEMATICAL MODELS AND METHODS IN APPLIED SCIENCES

Year of study

2016

Number

10

State

US

Pages from

373

Pages to

380

Pages count

8

Documents

BibTex


@article{BUT133164,
  author="Zdeněk {Kala}",
  title="Global Interval Sensitivity Analysis of Hermite Probability Density Function Percentiles",
  annote="The derivation of a Hermite probability density function whose parameters are mean value, standard deviation, skewness and kurtosis is presented in the article. Global interval sensitivity analysis capable of evaluating the effects of all combinations of the intervals of input parameters on the model output is presented. Global interval sensitivity analysis was used to identify the effects of uncertainties of the parameters of Hermite probability density function on its 5 percentile. A strong influence of the interaction between skewness and kurtosis, which would not have been identified using local sensitivity analysis, was identified. The global interval sensitivity analysis and Hermite probability density function presented here provide in combination with Monte Carlo simulation methods a very powerful tool for the probabilistic evaluation of structural reliability.",
  chapter="133164",
  howpublished="online",
  number="10",
  volume="2016",
  year="2016",
  month="october",
  pages="373--380",
  type="journal article - other"
}