Open AccessASSESSING EXTERNAL THREATS TO STRUCTURES USING LIMITED STATISTICAL DATA: AN APPROACH BASED ON DATA RESAMPLING
Author(s) -
Egidijus Rytas Vaidogas,
Virmantas Juocevičius
Publication year - 2007
Publication title -
technological and economic development of economy
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.634
H-Index - 47
eISSN - 2029-4921
pISSN - 2029-4913
DOI - 10.3846/13928619.2007.9637793
Subject(s) - resampling , representativeness heuristic , computer science , data mining , key (lock) , jackknife resampling , statistical hypothesis testing , sample size determination , confidence interval , econometrics , statistics , algorithm , mathematics , computer security , estimator
The paper deals with an estimation of risk to structures posed by extreme, dangerous phenomena called in brief the external threats. It is considered how to calculate risk values when a limited amount of data on actions imposed by these phenomena is available. The key methodology suggested in the paper for estimating the risk is the so‐called bootstrap resampling, known also as statistical or Efron's resampling. The paper presents a procedure allowing to apply the limited data to calculating bootstrap confidence intervals for probabilities of damage which can be caused by the actions. The application of the procedure is based on the assumption that the limited data can be expressed in the form of statistical sample which possesses the property of representativeness. It is discussed how to incorporate the confidence intervals in an expression of the risk induced by external threats. The proposed procedure can be viewed as a way of utilising limited and often very expensive data gained from experiments imitating occurrences of external threats. Findings described in this paper can be applied to design the structures for the so‐called accidental situations.