Volume 29, Issue 3

A new methodology that decomposes shocks into homoscedastic and heteroscedastic components is developed. This specification implies there exist linear combinations of heteroscedastic variables that eliminate heteroscedasticity; a property known as co-heteroscedasticity. The heteroscedastic part of the model uses a multivariate stochastic volatility inverse Wishart process. The resulting model is invariant to the ordering of the variables, which is shown to be important for volatility estimation. By incorporating testable co-heteroscedasticity restrictions, the specification allows estimation in moderately high-dimensions. The computational strategy uses a novel particle filter algorithm, a reparameterization that substantially improves algorithmic convergence and an alternating-order particle Gibbs that reduces the amount of particles needed for accurate estimation. An empirical application to a large Vector Autoregression (VAR) is provided, finding strong evidence for co-heteroscedasticity and that the new method outperforms some previously proposed methods in terms of forecasting at all horizons. It is also found that the structural monetary shock is 98.8 % homoscedastic, and that investment and the SP 500 index are nearly 100 % determined by fat tail heteroscedastic shocks. A Monte Carlo experiment illustrates that the new method estimates well the characteristics of approximate factor models with heteroscedastic errors.

This study examined extreme return spillovers and connectedness between crude oil (West Texas Intermediate), the Volatility Uncertainty Index (VIX), S&P 500, and six Latin American stock markets, namely, Argentina, Brazil, Chile, Colombia, Mexico, and Peru, using quantile connectedness. This approach allowed for a nuanced investigation of connectedness and added to the understanding the integration between these markets. The results indicated that the S&P 500 market was a full sender of spillover in the whole sample of the quantiles, when, to the contrary, the oil market was the highest receiver. The total spillovers were more intense during extreme quantiles, with swings between transmission and reception for VIX, Colombia, Mexico, and Peru. In addition, when the market turned to operate during bullish conditions, the VIX became a strong sender of spillover. Furthermore, an intense spillover was observed only in the lower and upper quantiles, and the spillover was sharper for the extreme upper quantile.

In this paper we show that a regression-based approach can be used to estimate generalised entropy and Atkinson inequality indices and their associated standard errors. The applicability of this approach is demonstrated using the health expenditure data from the United States (US) medical expenditure panel survey (MEPS).

Homogeneity identification of panel data models has been popular in the literature in recent years. Most of the existing works only focus on the complete data case. This paper considers a functional-coefficient quantile regression model for panel data with homogeneity when its response variables are subject to censoring. In particular, we consider a more general censoring framework, i.e. different types of censoring are allowed to occur in the model simultaneously. For this, a “three-stage” method is proposed, which includes the preliminary estimation of subject-specific function coefficients based on data augmentation, the identification of group structure over subjects by clustering, and post-grouping estimation of function coefficients. Simulation studies considering the left-, right-, and double-censored data, are carried out to verify the finite-sample properties of the proposed method. Simulation results show that our method gives comparable performance to the complete data case. The application to the bank stock data further illustrates the practical advantages of this method.

The two-stage estimator is often more tractable when there are nuisance parameters that can be separately estimated and plugged into an objective function. The joint estimator tends to bear the higher computational cost since it estimates all parameters in one stage by optimizing the sum of objective functions used in the two stages. It is well-known that the joint estimator is asymptotically more efficient than the two-stage estimator if the objective function is the true log-likelihood. When the objective function is not the true log-likelihood, I show that the relative asymptotic efficiency of the joint estimator still holds under a finite number of testable moment conditions. The implications of the main result on models based on quasi-limited information likelihoods are discussed.

China launched its first crude oil futures on 18th March 2018, with the primary objective of introducing its own hedging instrument in China. Our study responds to this objective by investigating the dependence structure based on time-variant and time-invariant copula, connectedness in various market conditions, and hedging effectiveness of oil futures with eleven GICS sectors in China using daily returns data from 19th June 2019 to 24th February 2024. Findings suggest that time-varying copula is a best fit for all the GICS sectors in China except for the Energy. Similarly, the results of time-varying (TVP-VAR) connectedness indicate that the oil futures are the net receiver of shocks in total, short (1–5 days), and long-time (5 to infinity) spillover. The oil futures can better offset portfolio losses during the down market than normal and up markets. Lastly, based on TVP-VAR, the hedging ratio, optimum portfolio weights, and hedge effectiveness are calculated. We find a positive hedge ratio across various pairs of assets, which reduces the assets’ volatility. Moreover, hedge ratios and optimum portfolio weights are non-constant over time. Our results significantly contribute to the scant literature on Chinese crude oil futures and have implications for investors and policymakers.