We propose methods to improve the forecasts from generalized autoregressive score (GAS) models (Creal, et al., 2013; Harvey, 2013) by localizing their parameters using decision trees and random forests. These methods avoid the curse of dimensionality faced by kernel-based approaches, and allow one...

The episodes of stock market crises in Europe and the U.S.A. since the year 2000, and the fragility of the international stock markets, have sparked the interest of researchers in understanding and in modeling the markets’ rising volatilities in order to prevent against crises. Portfolio...

The episodes of stock market crises in Europe and the U.S.A. since the year 2000, and the fragility of the New Technology sector after the explosion of the speculative bubble, have sparked the interest of researchers in understanding and in modeling this market’s high volatility to prevent...

We nowcast and forecast Austrian economic activity, namely real gross domestic product (GDP), consumption and investment, which are available at a quarterly frequency. While nowcasting uses data up to (and including) the quarter to be predicted, forecasting uses only data up to the previous...

Statistics Netherlands uses a state space model to estimate the Dutch unemployment by using monthly series about the labour force surveys (LFS). More accurate estimates of this variable can be obtained by including auxiliary information in the model, such as the univariate administrative series...

We prove that the diffusion limit of Real-Time GARCH (RT-GARCH) exists if we introduce an auxiliary process to replace the squared return in the volatility process. The volatility of the diffusion follows an Ornstein-Uhlenbeck-type process which fails to be positive with probability one unless...

Risk management is critical in decision-making, and mean-variance (MV) trade-off is one of the most common criteria. However, in reinforcement learning (RL) under a dynamic environment, MV control is not as easy as that under a static environment owing to computational difficulties. For MV...

We propose a new class of conditional heteroskedasticity in the volatility (CH-V) models which allows for time-varying volatility of volatility in the volatility of asset returns. This class nests a variety of GARCH-type models and the SHARV model of Ding (2021b). CH-V models can be seen as a...

This paper estimates univariate and multivariate conditional volatility and conditional correlation models of spot, forward and futures returns from three major benchmarks of international crude oil markets, namely Brent, WTI and Dubai, to aid in risk diversification. Conditional correlations...