Inference for Adaptive Time Series Models: Stochastic Volatility and Conditionally Gaussian State Space Form.
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 algoritms for this type of model are ineffective, but that this problem can be removed by reparameteris...
| Main Authors: | , |
|---|---|
| Format: | Working paper |
| Language: | English |
| Published: |
Nuffield College (University of Oxford)
2004
|
| _version_ | 1826302676949794816 |
|---|---|
| author | Bos, C Shephard, N |
| author_facet | Bos, C Shephard, N |
| author_sort | Bos, C |
| collection | OXFORD |
| description | 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 algoritms 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 literature. We also develop an effective particle filter for this model which is useful to assess the fit of the model. |
| first_indexed | 2024-03-07T05:51:11Z |
| format | Working paper |
| id | oxford-uuid:e8eec2bf-28c6-4e09-95ff-3a8369e261b9 |
| institution | University of Oxford |
| language | English |
| last_indexed | 2024-03-07T05:51:11Z |
| publishDate | 2004 |
| publisher | Nuffield College (University of Oxford) |
| record_format | dspace |
| spelling | oxford-uuid:e8eec2bf-28c6-4e09-95ff-3a8369e261b92022-03-27T10:50:26ZInference for Adaptive Time Series Models: Stochastic Volatility and Conditionally Gaussian State Space Form.Working paperhttp://purl.org/coar/resource_type/c_8042uuid:e8eec2bf-28c6-4e09-95ff-3a8369e261b9EnglishDepartment of Economics - ePrintsNuffield College (University of Oxford)2004Bos, CShephard, NIn 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 algoritms 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 literature. We also develop an effective particle filter for this model which is useful to assess the fit of the model. |
| spellingShingle | Bos, C Shephard, N Inference for Adaptive Time Series Models: Stochastic Volatility and Conditionally Gaussian State Space Form. |
| title | Inference for Adaptive Time Series Models: Stochastic Volatility and Conditionally Gaussian State Space Form. |
| title_full | Inference for Adaptive Time Series Models: Stochastic Volatility and Conditionally Gaussian State Space Form. |
| title_fullStr | Inference for Adaptive Time Series Models: Stochastic Volatility and Conditionally Gaussian State Space Form. |
| title_full_unstemmed | Inference for Adaptive Time Series Models: Stochastic Volatility and Conditionally Gaussian State Space Form. |
| title_short | Inference for Adaptive Time Series Models: Stochastic Volatility and Conditionally Gaussian State Space Form. |
| title_sort | inference for adaptive time series models stochastic volatility and conditionally gaussian state space form |
| work_keys_str_mv | AT bosc inferenceforadaptivetimeseriesmodelsstochasticvolatilityandconditionallygaussianstatespaceform AT shephardn inferenceforadaptivetimeseriesmodelsstochasticvolatilityandconditionallygaussianstatespaceform |