Deep Neural Network Algorithms for Parabolic PIDEs and Applications in Insurance and Finance

Rüdiger Frey; Verena Köck

doi:10.3390/computation10110201

Deep Neural Network Algorithms for Parabolic PIDEs and Applications in Insurance and Finance

Rüdiger Frey, Verena Köck^*

^*Corresponding author for this work

Institute for Statistics and Mathematics

Publication: Scientific journal › Journal article › peer-review

Abstract

In this paper we study deep neural network algorithms for solving inear and semilinear parabolic partial integro-differential equations with boundary conditions in high dimension. Our method can be considered as an extension of the deep splitting method for PDEs to equations with non-local terms. To show the viability of our approach, we discuss several case studies from insurance and finance.

Original language	English
Article number	201
Journal	Computation
Volume	10
Issue number	11
DOIs	https://doi.org/10.3390/computation10110201
Publication status	Published - Nov 2022

Bibliographical note

Publisher Copyright:
© 2022 by the authors.

Keywords

deep neural networks
finance
insurance
machine earning
parabolic partial integro-differential equations

Access to Document

10.3390/computation10110201

Cite this

@article{0cf0c1aeab224dfa8bb70db04136e78a,

title = "Deep Neural Network Algorithms for Parabolic PIDEs and Applications in Insurance and Finance",

abstract = "In this paper we study deep neural network algorithms for solving inear and semilinear parabolic partial integro-differential equations with boundary conditions in high dimension. Our method can be considered as an extension of the deep splitting method for PDEs to equations with non-local terms. To show the viability of our approach, we discuss several case studies from insurance and finance.",

keywords = "deep neural networks, finance, insurance, machine earning, parabolic partial integro-differential equations",

author = "R{\"u}diger Frey and Verena K{\"o}ck",

note = "Publisher Copyright: {\textcopyright} 2022 by the authors.",

year = "2022",

month = nov,

doi = "10.3390/computation10110201",

language = "English",

volume = "10",

journal = "Computation",

issn = "2079-3197",

publisher = "Multidisciplinary Digital Publishing Institute (MDPI)",

number = "11",

}

TY - JOUR

T1 - Deep Neural Network Algorithms for Parabolic PIDEs and Applications in Insurance and Finance

AU - Frey, Rüdiger

AU - Köck, Verena

PY - 2022/11

Y1 - 2022/11

N2 - In this paper we study deep neural network algorithms for solving inear and semilinear parabolic partial integro-differential equations with boundary conditions in high dimension. Our method can be considered as an extension of the deep splitting method for PDEs to equations with non-local terms. To show the viability of our approach, we discuss several case studies from insurance and finance.

AB - In this paper we study deep neural network algorithms for solving inear and semilinear parabolic partial integro-differential equations with boundary conditions in high dimension. Our method can be considered as an extension of the deep splitting method for PDEs to equations with non-local terms. To show the viability of our approach, we discuss several case studies from insurance and finance.

KW - deep neural networks

KW - finance

KW - insurance

KW - machine earning

KW - parabolic partial integro-differential equations

UR - https://www.scopus.com/pages/publications/85141566717

U2 - 10.3390/computation10110201

DO - 10.3390/computation10110201

M3 - Journal article

AN - SCOPUS:85141566717

SN - 2079-3197

VL - 10

JO - Computation

JF - Computation

IS - 11

M1 - 201

ER -

Deep Neural Network Algorithms for Parabolic PIDEs and Applications in Insurance and Finance

Abstract

Bibliographical note

Keywords

Access to Document

Other files and links

Cite this