|
This article is cited in 3 scientific papers (total in 3 papers)
Importance sampling for simulation of large and moderate deviation probabilities of tests and estimators
M. S. Ermakov Institute of Problems of Mechanical Engineering, Russian Academy of Sciences
Abstract:
Using the machinery of large and moderate deviations theory of empirical probability measures, we study effective importance sampling for simulation of large and moderate deviation probabilities of tests and estimators. The computational burden of efficient importance sampling does not have the exponential growth as in the straightforward simulation. The results are implemented in a simulation of moderate deviation probabilities of tests of omega-squared type.
Keywords:
importance sampling, large deviations, moderate deviations, empirical measure.
Received: 19.04.2004 Revised: 24.10.2005
Citation:
M. S. Ermakov, “Importance sampling for simulation of large and moderate deviation probabilities of tests and estimators”, Teor. Veroyatnost. i Primenen., 51:2 (2006), 319–332; Theory Probab. Appl., 51:2 (2007), 279–290
Linking options:
https://www.mathnet.ru/eng/tvp56https://doi.org/10.4213/tvp56 https://www.mathnet.ru/eng/tvp/v51/i2/p319
|
|