The semiparametric asymmetric stochastic volatility model with time-varying parameters: The case of US inflation

- A semiparametric asymmetric stochastic volatility model with time-varying parameters is considered.
- An efficient Markov Chain Monte Carlo estimation algorithm is developed.
- The proposed model is applied to inflation modeling.
- The proposed model shows positive correlation between inflation and volatility.
- The proposed model forecasts better that competing models.

We propose a semiparametric extension of the time-varying parameter regression model with asymmetric stochastic volatility. For parameter estimation we use Bayesian methods. We illustrate our methods with an application to US inflation.