Inference for Adaptive Time Series Models: Stochastic Volatility and Conditionally Gaussian State Space form

Charles S. Bos, Neil Shephard

Research output: Working paperProfessional

Abstract

In this paper we replace the Gaussian errors in the standard Gaussian, linear state space model with stochastic volatility processes. This is called a GSSF-SV model. We show that conventional MCMC algorithms for this type of model are ineffective, but that this problem can be removed by reparameterising the model. We illustrate our results on an example from financial economics and one from the nonparametric regression model. We also develop an effective particle filter for this model which is useful to assess the fit of the model.
Original languageEnglish
Place of PublicationAmsterdam
PublisherTinbergen Instituut
Publication statusPublished - 2004

Publication series

NameDiscussion paper TI
No.04-015/4

Fingerprint

Stochastic Volatility
Space Form
Time Series Models
State Space
MCMC Algorithm
Model
Nonparametric Model
Particle Filter
State-space Model
Nonparametric Regression
Linear Space
Linear Model
Regression Model
Economics

Cite this

Bos, C. S., & Shephard, N. (2004). Inference for Adaptive Time Series Models: Stochastic Volatility and Conditionally Gaussian State Space form. (Discussion paper TI; No. 04-015/4). Amsterdam: Tinbergen Instituut.
Bos, Charles S. ; Shephard, Neil. / Inference for Adaptive Time Series Models: Stochastic Volatility and Conditionally Gaussian State Space form. Amsterdam : Tinbergen Instituut, 2004. (Discussion paper TI; 04-015/4).
@techreport{204b2f4b2d52467c8d8768e50d909b7b,
title = "Inference for Adaptive Time Series Models: Stochastic Volatility and Conditionally Gaussian State Space form",
abstract = "In this paper we replace the Gaussian errors in the standard Gaussian, linear state space model with stochastic volatility processes. This is called a GSSF-SV model. We show that conventional MCMC algorithms for this type of model are ineffective, but that this problem can be removed by reparameterising the model. We illustrate our results on an example from financial economics and one from the nonparametric regression model. We also develop an effective particle filter for this model which is useful to assess the fit of the model.",
author = "Bos, {Charles S.} and Neil Shephard",
year = "2004",
language = "English",
series = "Discussion paper TI",
publisher = "Tinbergen Instituut",
number = "04-015/4",
type = "WorkingPaper",
institution = "Tinbergen Instituut",

}

Bos, CS & Shephard, N 2004 'Inference for Adaptive Time Series Models: Stochastic Volatility and Conditionally Gaussian State Space form' Discussion paper TI, no. 04-015/4, Tinbergen Instituut, Amsterdam.

Inference for Adaptive Time Series Models: Stochastic Volatility and Conditionally Gaussian State Space form. / Bos, Charles S.; Shephard, Neil.

Amsterdam : Tinbergen Instituut, 2004. (Discussion paper TI; No. 04-015/4).

Research output: Working paperProfessional

TY - UNPB

T1 - Inference for Adaptive Time Series Models: Stochastic Volatility and Conditionally Gaussian State Space form

AU - Bos, Charles S.

AU - Shephard, Neil

PY - 2004

Y1 - 2004

N2 - In this paper we replace the Gaussian errors in the standard Gaussian, linear state space model with stochastic volatility processes. This is called a GSSF-SV model. We show that conventional MCMC algorithms for this type of model are ineffective, but that this problem can be removed by reparameterising the model. We illustrate our results on an example from financial economics and one from the nonparametric regression model. We also develop an effective particle filter for this model which is useful to assess the fit of the model.

AB - In this paper we replace the Gaussian errors in the standard Gaussian, linear state space model with stochastic volatility processes. This is called a GSSF-SV model. We show that conventional MCMC algorithms for this type of model are ineffective, but that this problem can be removed by reparameterising the model. We illustrate our results on an example from financial economics and one from the nonparametric regression model. We also develop an effective particle filter for this model which is useful to assess the fit of the model.

M3 - Working paper

T3 - Discussion paper TI

BT - Inference for Adaptive Time Series Models: Stochastic Volatility and Conditionally Gaussian State Space form

PB - Tinbergen Instituut

CY - Amsterdam

ER -

Bos CS, Shephard N. Inference for Adaptive Time Series Models: Stochastic Volatility and Conditionally Gaussian State Space form. Amsterdam: Tinbergen Instituut. 2004. (Discussion paper TI; 04-015/4).