Filtering and Parameter Estimation of Partially Observed Diffusion Processes Using Gaussian RBFs

Josef Höök; Elisabeth Larsson; Erik Lindström; Lina von Sydow

Filtering and Parameter Estimation of Partially Observed Diffusion Processes Using Gaussian RBFs

Josef Höök, Elisabeth Larsson, Erik Lindström, Lina von Sydow

Forskningsoutput: Tidskriftsbidrag › Publicerat konferensabstract › Peer review

Originalspråk	engelska
Tidskrift	[Publication information missing]
Status	Published - 2014
Evenemang	2014 SIAM Conference on Financial Mathematics and Engineering - Chicago, Illinois, USA Varaktighet: 2014 nov. 13 → 2014 nov. 15

Bibliografisk information

Asset prices can be modeled as stochastic diffusion pro-
cesses involving a number of parameters. Based on market
observations, these parameters can be estimated. Prices
are not uniquely determined due to the ask-bid spread. We
model the spread as additive noise, and show that Gaussian
radial basis functions (RBFs), leads to a convenient math-
ematical representation. Furthermore, substantial parts of
the computations can be performed analytically if RBFs
are used for approximating transition densities.

Ämnesklassifikation (UKÄ)

Sannolikhetsteori och statistik

Tillgång till dokument

http://www.siam.org/meetings/fm14/fm14_abstracts.pdf

Citera det här

@article{8abd00db48564494bfcdda65b3703bb8,

title = "Filtering and Parameter Estimation of Partially Observed Diffusion Processes Using Gaussian RBFs",

keywords = "RBF, Filtering, Fokker-Planck equation",

author = "Josef H{\"o}{\"o}k and Elisabeth Larsson and Erik Lindstr{\"o}m and {von Sydow}, Lina",

note = "Asset prices can be modeled as stochastic diffusion pro- cesses involving a number of parameters. Based on market observations, these parameters can be estimated. Prices are not uniquely determined due to the ask-bid spread. We model the spread as additive noise, and show that Gaussian radial basis functions (RBFs), leads to a convenient math- ematical representation. Furthermore, substantial parts of the computations can be performed analytically if RBFs are used for approximating transition densities.; 2014 SIAM Conference on Financial Mathematics and Engineering ; Conference date: 13-11-2014 Through 15-11-2014",

year = "2014",

language = "English",

journal = "[Publication information missing]",

}

TY - JOUR

T1 - Filtering and Parameter Estimation of Partially Observed Diffusion Processes Using Gaussian RBFs

AU - Höök, Josef

AU - Larsson, Elisabeth

AU - Lindström, Erik

AU - von Sydow, Lina

N1 - Asset prices can be modeled as stochastic diffusion pro- cesses involving a number of parameters. Based on market observations, these parameters can be estimated. Prices are not uniquely determined due to the ask-bid spread. We model the spread as additive noise, and show that Gaussian radial basis functions (RBFs), leads to a convenient math- ematical representation. Furthermore, substantial parts of the computations can be performed analytically if RBFs are used for approximating transition densities.

PY - 2014

Y1 - 2014

KW - RBF

KW - Filtering

KW - Fokker-Planck equation

M3 - Published meeting abstract

JO - [Publication information missing]

JF - [Publication information missing]

T2 - 2014 SIAM Conference on Financial Mathematics and Engineering

Y2 - 13 November 2014 through 15 November 2014

ER -