Research output: Working paper/Preprint › Working paper
A Generalized Schwartz Model for Energy Spot Prices - Estimation using a Particle MCMC Method. / Lunde, Asger; Brix, Anne Floor; Wei, Wei.
Aarhus : Institut for Økonomi, Aarhus Universitet, 2015.Research output: Working paper/Preprint › Working paper
}
TY - UNPB
T1 - A Generalized Schwartz Model for Energy Spot Prices - Estimation using a Particle MCMC Method
AU - Lunde, Asger
AU - Brix, Anne Floor
AU - Wei, Wei
PY - 2015/10/12
Y1 - 2015/10/12
N2 - We propose an energy spot price model featuring a two-factor price process and a two-component stochastic volatility process. The first factor in the price process captures the normal variations; the second accounts for spikes. The two-component volatility allows for a flexible autocorrelation structure. Instead of using various filtering techniques for splitting the two factors, as often found in the literature, we estimate the model in one step using an adaptive MCMC method with a Rao-Blackwellized particle filter. We fit the model to UK natural gas spot prices and investigate the importance of spikes and stochastic volatility. We find that the inclusion of stochastic volatility is crucial and that it strongly impacts the jump intensity in the spike process. Furthermore, our estimation method enables us to consider both continuous and purely jump-driven volatility processes, and thereby assess if the volatility specification also influences the spike process and the overall model fit.
AB - We propose an energy spot price model featuring a two-factor price process and a two-component stochastic volatility process. The first factor in the price process captures the normal variations; the second accounts for spikes. The two-component volatility allows for a flexible autocorrelation structure. Instead of using various filtering techniques for splitting the two factors, as often found in the literature, we estimate the model in one step using an adaptive MCMC method with a Rao-Blackwellized particle filter. We fit the model to UK natural gas spot prices and investigate the importance of spikes and stochastic volatility. We find that the inclusion of stochastic volatility is crucial and that it strongly impacts the jump intensity in the spike process. Furthermore, our estimation method enables us to consider both continuous and purely jump-driven volatility processes, and thereby assess if the volatility specification also influences the spike process and the overall model fit.
KW - Energy prices, Multi-factor model, Particle filters, MCMC, Stochastic volatility
M3 - Working paper
T3 - CREATES Research Papers
BT - A Generalized Schwartz Model for Energy Spot Prices - Estimation using a Particle MCMC Method
PB - Institut for Økonomi, Aarhus Universitet
CY - Aarhus
ER -