An inverse gamma activity time process with noninteger parameters and a self-similar limit

Richard Finlay; Eugene Seneta; Dingcheng Wang

doi:10.1239/jap/1339878797

An inverse gamma activity time process with noninteger parameters and a self-similar limit

Richard Finlay^*, Eugene Seneta, Dingcheng Wang

^*Corresponding author for this work

Research output: Contribution to journal › Article › peer-review

5 Citations (Scopus)

Abstract

We construct a process with inverse gamma increments and an asymptotically self-similar limit. This construction supports the use of long-range-dependent t subordinator models for actual financial data as advocated in Heyde and Leonenko (2005), in that it allows for noninteger-valued model parameters, as is found empirically in model estimation from data.

Original language	English
Pages (from-to)	441-450
Number of pages	10
Journal	Journal of Applied Probability
Volume	49
Issue number	2
DOIs	https://doi.org/10.1239/jap/1339878797
Publication status	Published - Jun 2012

Access to Document

10.1239/jap/1339878797

Cite this

@article{06dda15f8ddb4c3883a2127dede46664,

title = "An inverse gamma activity time process with noninteger parameters and a self-similar limit",

abstract = "We construct a process with inverse gamma increments and an asymptotically self-similar limit. This construction supports the use of long-range-dependent t subordinator models for actual financial data as advocated in Heyde and Leonenko (2005), in that it allows for noninteger-valued model parameters, as is found empirically in model estimation from data.",

keywords = "Inverse gamma process, Long-range dependence, Self similarity, Subordinator model, T-distribution",

author = "Richard Finlay and Eugene Seneta and Dingcheng Wang",

year = "2012",

month = jun,

doi = "10.1239/jap/1339878797",

language = "English",

volume = "49",

pages = "441--450",

journal = "Journal of Applied Probability",

issn = "0021-9002",

publisher = "Cambridge University Press",

number = "2",

}

TY - JOUR

T1 - An inverse gamma activity time process with noninteger parameters and a self-similar limit

AU - Finlay, Richard

AU - Seneta, Eugene

AU - Wang, Dingcheng

PY - 2012/6

Y1 - 2012/6

N2 - We construct a process with inverse gamma increments and an asymptotically self-similar limit. This construction supports the use of long-range-dependent t subordinator models for actual financial data as advocated in Heyde and Leonenko (2005), in that it allows for noninteger-valued model parameters, as is found empirically in model estimation from data.

AB - We construct a process with inverse gamma increments and an asymptotically self-similar limit. This construction supports the use of long-range-dependent t subordinator models for actual financial data as advocated in Heyde and Leonenko (2005), in that it allows for noninteger-valued model parameters, as is found empirically in model estimation from data.

KW - Inverse gamma process

KW - Long-range dependence

KW - Self similarity

KW - Subordinator model

KW - T-distribution

UR - http://www.scopus.com/inward/record.url?scp=84872231266&partnerID=8YFLogxK

U2 - 10.1239/jap/1339878797

DO - 10.1239/jap/1339878797

M3 - Article

SN - 0021-9002

VL - 49

SP - 441

EP - 450

JO - Journal of Applied Probability

JF - Journal of Applied Probability

IS - 2

ER -

An inverse gamma activity time process with noninteger parameters and a self-similar limit

Abstract

Access to Document

Other files and links

Fingerprint

Cite this