Zero-crossing statistics for non-Markovian time series
(2018) In Physical Review E 97(3).- Abstract
In applications spanning from image analysis and speech recognition to energy dissipation in turbulence and time-to failure of fatigued materials, researchers and engineers want to calculate how often a stochastic observable crosses a specific level, such as zero. At first glance this problem looks simple, but it is in fact theoretically very challenging, and therefore few exact results exist. One exception is the celebrated Rice formula that gives the mean number of zero crossings in a fixed time interval of a zero-mean Gaussian stationary process. In this study we use the so-called independent interval approximation to go beyond Rice's result and derive analytic expressions for all higher-order zero-crossing cumulants and moments. Our... (More)
In applications spanning from image analysis and speech recognition to energy dissipation in turbulence and time-to failure of fatigued materials, researchers and engineers want to calculate how often a stochastic observable crosses a specific level, such as zero. At first glance this problem looks simple, but it is in fact theoretically very challenging, and therefore few exact results exist. One exception is the celebrated Rice formula that gives the mean number of zero crossings in a fixed time interval of a zero-mean Gaussian stationary process. In this study we use the so-called independent interval approximation to go beyond Rice's result and derive analytic expressions for all higher-order zero-crossing cumulants and moments. Our results agree well with simulations for the non-Markovian autoregressive model.
(Less)
- author
- Nyberg, Markus ; Lizana, Ludvig and Ambjörnsson, Tobias LU
- organization
- publishing date
- 2018-03-14
- type
- Contribution to journal
- publication status
- published
- subject
- in
- Physical Review E
- volume
- 97
- issue
- 3
- article number
- 032114
- publisher
- American Physical Society
- external identifiers
-
- pmid:29776037
- scopus:85044118092
- ISSN
- 2470-0045
- DOI
- 10.1103/PhysRevE.97.032114
- language
- English
- LU publication?
- yes
- id
- abb41c8e-1e14-4fa2-b924-b7d628e9305a
- date added to LUP
- 2018-04-04 13:30:31
- date last changed
- 2024-04-15 04:45:02
@article{abb41c8e-1e14-4fa2-b924-b7d628e9305a,
abstract = {{<p>In applications spanning from image analysis and speech recognition to energy dissipation in turbulence and time-to failure of fatigued materials, researchers and engineers want to calculate how often a stochastic observable crosses a specific level, such as zero. At first glance this problem looks simple, but it is in fact theoretically very challenging, and therefore few exact results exist. One exception is the celebrated Rice formula that gives the mean number of zero crossings in a fixed time interval of a zero-mean Gaussian stationary process. In this study we use the so-called independent interval approximation to go beyond Rice's result and derive analytic expressions for all higher-order zero-crossing cumulants and moments. Our results agree well with simulations for the non-Markovian autoregressive model.</p>}},
author = {{Nyberg, Markus and Lizana, Ludvig and Ambjörnsson, Tobias}},
issn = {{2470-0045}},
language = {{eng}},
month = {{03}},
number = {{3}},
publisher = {{American Physical Society}},
series = {{Physical Review E}},
title = {{Zero-crossing statistics for non-Markovian time series}},
url = {{http://dx.doi.org/10.1103/PhysRevE.97.032114}},
doi = {{10.1103/PhysRevE.97.032114}},
volume = {{97}},
year = {{2018}},
}