Loading...
Please wait, while we are loading the content...
Similar Documents
Autoregressive Models with Time-Varying Dependence and Stochastic Volatility: An Application to Inflation Forecasting
| Content Provider | Semantic Scholar |
|---|---|
| Author | Turatti, Douglas Eduardo Koopman, Siem Jan |
| Copyright Year | 2017 |
| Abstract | This article proposes an efficient estimation procedure for models containing multiple linear and non-linear time-varying parameters and stationarity restrictions. The methodology is based on a multivariate extension of the numerically accelerated importance sampling together with a Rao-Blackwellization step to construct a highly efficient estimation procedure. The approach compares favorably to the particle filter for likelihood evaluation. We base our analysis on a time-varying autogressive model with stochastic volatility. The proposed model is used to forecast U.S. C.P.I. inflation and showed superior results with respect to benchmark models. JEL classification: C32; C63; E37 |
| File Format | PDF HTM / HTML |
| Alternate Webpage(s) | https://editorialexpress.com/cgi-bin/conference/download.cgi?db_name=SAEe2017&paper_id=550 |
| Language | English |
| Access Restriction | Open |
| Content Type | Text |
| Resource Type | Article |
