Finding the embedding dimension and variable dependencies in time series
(1994) In Neural Computation 6(3). p.509-520- Abstract
- We present a general method, the δ-test, which establishes functional dependencies given a sequence of measurements. The approach is based on calculating conditional probabilities from vector component distances. Imposing the requirement of continuity of the underlying function, the obtained values of the conditional probabilities carry information on the embedding dimension and variable dependencies. The power of the method is illustrated on synthetic time-series with different time-lag dependencies and noise levels and on the sunspot data. The virtue of the method for preprocessing data in the context of feedforward neural networks is demonstrated. Also, its applicability for tracking residual errors in output units is stressed.
Please use this url to cite or link to this publication:
https://lup.lub.lu.se/record/56b85c76-d63b-4209-b417-ba5cfdc4ef81
- author
- Pi, Hong and Peterson, Carsten LU
- organization
- publishing date
- 1994
- type
- Contribution to journal
- publication status
- published
- subject
- in
- Neural Computation
- volume
- 6
- issue
- 3
- pages
- 11 pages
- publisher
- MIT Press
- ISSN
- 1530-888X
- DOI
- 10.1162/neco.1994.6.3.509
- language
- English
- LU publication?
- yes
- id
- 56b85c76-d63b-4209-b417-ba5cfdc4ef81
- date added to LUP
- 2019-05-31 16:29:49
- date last changed
- 2019-08-07 10:38:52
@article{56b85c76-d63b-4209-b417-ba5cfdc4ef81, abstract = {{We present a general method, the δ-test, which establishes functional dependencies given a sequence of measurements. The approach is based on calculating conditional probabilities from vector component distances. Imposing the requirement of continuity of the underlying function, the obtained values of the conditional probabilities carry information on the embedding dimension and variable dependencies. The power of the method is illustrated on synthetic time-series with different time-lag dependencies and noise levels and on the sunspot data. The virtue of the method for preprocessing data in the context of feedforward neural networks is demonstrated. Also, its applicability for tracking residual errors in output units is stressed.}}, author = {{Pi, Hong and Peterson, Carsten}}, issn = {{1530-888X}}, language = {{eng}}, number = {{3}}, pages = {{509--520}}, publisher = {{MIT Press}}, series = {{Neural Computation}}, title = {{Finding the embedding dimension and variable dependencies in time series}}, url = {{http://dx.doi.org/10.1162/neco.1994.6.3.509}}, doi = {{10.1162/neco.1994.6.3.509}}, volume = {{6}}, year = {{1994}}, }