Asymmetric stable stochastic volatility models: estimation, filtering, and forecasting
本文提出一种非对称稳定误差分布的随机波动率模型，并采用间接推断和极值蒙特卡洛方法解决参数估计、波动率滤波和预测问题，对金融资产（如比特币）的波动率建模有参考价值。
This article considers a stochastic volatility model featuring an asymmetric stable error distribution and a novel way of accounting for the leverage effect. We adopt simulation‐based methods to address key challenges in parameter estimation, the filtering of time‐varying volatility, and volatility forecasting. More specifically, we make use of the indirect inference method for estimating the static parameters, while the latent volatility is extracted using the extremum Monte Carlo method. Both parameter estimation and volatility extraction are easily adapted to other model specifications, such as those based on other error distributions or on other dynamic processes for volatility. Illustrations are presented for a simulated dataset and for an empirical dataset of daily Bitcoin returns.