An Efficient Sequential Learning Algorithm in Regime-Switching Environments
Preprint
- 31 January 2018
- preprint
- Published by Elsevier in SSRN Electronic Journal
Abstract
We provide a noble approach of estimating a regime-switching nonlinear and non-Gaussian state space model based on a particle learning scheme. Particularly, we extend the particle learning method in Liu and West (2001) by constructing a new proposal distribution for the latent regime index variable that incorporates all available information contained in the current and past observations. The Monte Carlo simulation result implies that our approach categorically outperforms a popular existing algorithm. For an empirical illustration, the proposed algorithm is used to analyze the underlying dynamics of the U.S. excess stock return.Keywords
This publication has 17 references indexed in Scilit:
- Particle Learning and SmoothingStatistical Science, 2010
- Simulation-based sequential analysis of Markov switching stochastic volatility modelsComputational Statistics & Data Analysis, 2007
- There is a risk-return trade-off after allJournal of Financial Economics, 2005
- On the relationship between the conditional mean and volatility of stock returns: A latent VAR approachJournal of Financial Economics, 2004
- Combined Parameter and State Estimation in Simulation-Based Filtering2001
- Bayes FactorsJournal of the American Statistical Association, 1995
- Dynamic linear models with Markov-switchingJournal of Econometrics, 1994
- On the Relation between the Expected Value and the Volatility of the Nominal Excess Return on StocksThe Journal of Finance, 1993
- Novel approach to nonlinear/non-Gaussian Bayesian state estimationIEE Proceedings F Radar and Signal Processing, 1993
- The stochastic behavior of common stock variances Value, leverage and interest rate effectsJournal of Financial Economics, 1982