A method for density estimation based on expectation identities

Joaquín Peralta; Claudia Loyola; Humberto Loguercio; Sergio Davis

doi:10.1063/1.4985376

A method for density estimation based on expectation identities

Joaquín Peralta, Claudia Loyola, Humberto Loguercio, Sergio Davis

Universidad Andrés Bello

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

1 Citation (Scopus)

Abstract

We present a simple and direct method for non-parametric estimation of a one-dimensional probability density, based on the application of the recent conjugate variables theorem. The method expands the logarithm of the probability density ln P(x|I) in terms of a complete basis and numerically solves for the coefficients of the expansion using a linear system of equations. No Monte Carlo sampling is needed. We present preliminary results that show the practical usefulness of the method for modeling statistical data.

Original language	English
Title of host publication	Bayesian Inference and Maximum Entropy Methods in Science and Engineering
Subtitle of host publication	Proceedings of the 36th International Workshop on Bayesian Inference and Maximum Entropy Methods in Science and Engineering, MaxEnt 2016
Publisher	American Institute of Physics Inc.
Volume	1853
ISBN (Electronic)	9780735415270
DOIs	https://doi.org/10.1063/1.4985376
Publication status	Published - 9 Jun 2017
Event	36th International Workshop on Bayesian Inference and Maximum Entropy Methods in Science and Engineering, MaxEnt 2016 - Ghent, Belgium Duration: 10 Jul 2016 → 15 Jul 2016

Conference

Conference	36th International Workshop on Bayesian Inference and Maximum Entropy Methods in Science and Engineering, MaxEnt 2016
Country/Territory	Belgium
City	Ghent
Period	10/07/16 → 15/07/16

ASJC Scopus subject areas

General Physics and Astronomy

Access to Document

10.1063/1.4985376

Cite this

Peralta, J., Loyola, C., Loguercio, H., & Davis, S. (2017). A method for density estimation based on expectation identities. In Bayesian Inference and Maximum Entropy Methods in Science and Engineering: Proceedings of the 36th International Workshop on Bayesian Inference and Maximum Entropy Methods in Science and Engineering, MaxEnt 2016 (Vol. 1853). Article 110001 American Institute of Physics Inc.. https://doi.org/10.1063/1.4985376

Peralta, Joaquín ; Loyola, Claudia ; Loguercio, Humberto et al. / A method for density estimation based on expectation identities. Bayesian Inference and Maximum Entropy Methods in Science and Engineering: Proceedings of the 36th International Workshop on Bayesian Inference and Maximum Entropy Methods in Science and Engineering, MaxEnt 2016. Vol. 1853 American Institute of Physics Inc., 2017.

@inproceedings{8c3da547a85e40d695e8ecd00aebd38c,

title = "A method for density estimation based on expectation identities",

abstract = "We present a simple and direct method for non-parametric estimation of a one-dimensional probability density, based on the application of the recent conjugate variables theorem. The method expands the logarithm of the probability density ln P(x|I) in terms of a complete basis and numerically solves for the coefficients of the expansion using a linear system of equations. No Monte Carlo sampling is needed. We present preliminary results that show the practical usefulness of the method for modeling statistical data.",

author = "Joaqu{\'i}n Peralta and Claudia Loyola and Humberto Loguercio and Sergio Davis",

year = "2017",

month = jun,

day = "9",

doi = "10.1063/1.4985376",

language = "English",

volume = "1853",

booktitle = "Bayesian Inference and Maximum Entropy Methods in Science and Engineering",

publisher = "American Institute of Physics Inc.",

note = "36th International Workshop on Bayesian Inference and Maximum Entropy Methods in Science and Engineering, MaxEnt 2016 ; Conference date: 10-07-2016 Through 15-07-2016",

}

Peralta, J , Loyola, C, Loguercio, H & Davis, S 2017, A method for density estimation based on expectation identities. in Bayesian Inference and Maximum Entropy Methods in Science and Engineering: Proceedings of the 36th International Workshop on Bayesian Inference and Maximum Entropy Methods in Science and Engineering, MaxEnt 2016. vol. 1853, 110001, American Institute of Physics Inc., 36th International Workshop on Bayesian Inference and Maximum Entropy Methods in Science and Engineering, MaxEnt 2016, Ghent, Belgium, 10/07/16. https://doi.org/10.1063/1.4985376

A method for density estimation based on expectation identities. / Peralta, Joaquín ; Loyola, Claudia; Loguercio, Humberto et al.
Bayesian Inference and Maximum Entropy Methods in Science and Engineering: Proceedings of the 36th International Workshop on Bayesian Inference and Maximum Entropy Methods in Science and Engineering, MaxEnt 2016. Vol. 1853 American Institute of Physics Inc., 2017. 110001.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - A method for density estimation based on expectation identities

AU - Peralta, Joaquín

AU - Loyola, Claudia

AU - Loguercio, Humberto

AU - Davis, Sergio

PY - 2017/6/9

Y1 - 2017/6/9

N2 - We present a simple and direct method for non-parametric estimation of a one-dimensional probability density, based on the application of the recent conjugate variables theorem. The method expands the logarithm of the probability density ln P(x|I) in terms of a complete basis and numerically solves for the coefficients of the expansion using a linear system of equations. No Monte Carlo sampling is needed. We present preliminary results that show the practical usefulness of the method for modeling statistical data.

AB - We present a simple and direct method for non-parametric estimation of a one-dimensional probability density, based on the application of the recent conjugate variables theorem. The method expands the logarithm of the probability density ln P(x|I) in terms of a complete basis and numerically solves for the coefficients of the expansion using a linear system of equations. No Monte Carlo sampling is needed. We present preliminary results that show the practical usefulness of the method for modeling statistical data.

UR - http://www.scopus.com/inward/record.url?scp=85021442130&partnerID=8YFLogxK

U2 - 10.1063/1.4985376

DO - 10.1063/1.4985376

M3 - Conference contribution

AN - SCOPUS:85021442130

VL - 1853

BT - Bayesian Inference and Maximum Entropy Methods in Science and Engineering

PB - American Institute of Physics Inc.

T2 - 36th International Workshop on Bayesian Inference and Maximum Entropy Methods in Science and Engineering, MaxEnt 2016

Y2 - 10 July 2016 through 15 July 2016

ER -

Peralta J , Loyola C, Loguercio H, Davis S. A method for density estimation based on expectation identities. In Bayesian Inference and Maximum Entropy Methods in Science and Engineering: Proceedings of the 36th International Workshop on Bayesian Inference and Maximum Entropy Methods in Science and Engineering, MaxEnt 2016. Vol. 1853. American Institute of Physics Inc. 2017. 110001 doi: 10.1063/1.4985376

A method for density estimation based on expectation identities

Abstract

Conference

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this