Volume 21, Issue 1 - Special issue: Recent developments of switching models for financial data / Guest editors: Gilles Dufrénot and Fredj Jawadi

The regime-switching Lévy model combines jump-diffusion under the form of a Lévy process, and Markov regime-switching where all parameters depend on the value of a continuous time Markov chain. We start by giving general stochastic results. Estimation is performed following a two-step procedure. The EM-algorithm is extended to this new class of jump-diffusion regime-switching models. Simulations are proposed, alongside an empirical application dedicated to the study of financial and commodity time series. When comparing the results with (i) non regime-switching models, and (ii) continuous regime-switching models (where the Lévy process is replaced by a classic Brownian motion), the Lévy regime-switching model outperforms other competitors.

This study proposes a new unit root test that allows for structural breaks in both the intercept and the slope, and adopts the residual augmented least squares (RALS) procedure to gain improved power when the error term follows a non-normal distribution. The new test using the RALS procedure is more powerful than the usual LM test which does not incorporate information on non-normal errors. Our test is free of nuisance parameters that indicate the locations of structural break. It is also free of the spurious rejection problem. Thus, the rejection of the null hypothesis can be considered as more accurate evidence of stationarity. We apply the new test on the recently extended Grilli and Yang index of 24 commodity series from 1900 to 2007. Our empirical findings provide significant evidence that primary commodity prices are stationary with one or two trend breaks. However, compared with past studies, our findings provide even weaker evidence to support the Prebisch-Singer hypothesis.

This paper studies the hypothesis of stock price comovements between the US market and three different regions [the G6, BRICS and MENA (Middle East North Africa)] during calm and crisis periods. It extends the study by (Chlibi, S., F. Jawadi, and M. Sellami. 2016. “Analyzing Heterogeneous Stock Price Comovements through Hybrid Approaches.” Open Economies Review 27: 541–559) and models stock price comovements in a nonlinear multivariate framework. In particular, we develop an empirical nonlinear specification to model heterogeneity in stock price comovements, based on the multivariate threshold cointegration framework of (Hansen, B. E., and B. Seo. 2002. “Testing For Two-Regime Threshold Cointegration in Vector Error Correction Models.” Journal of Econometrics 110: 293–318). Our findings point to the presence of significant threshold cointegration relationships that are activated per regime. Specification of these threshold effects in stock price comovements is particularly useful to help investors to rebalance and adjust their portfolios so as to optimize their investment and diversification strategies.

We examine model specification in regime-switching continuous-time diffusions for modeling S&P 500 Volatility Index (VIX). Our investigation is carried out under two nonlinear diffusion frameworks, the NLDCEV and the CIRCEV frameworks, and our focus is on the nonlinearity in regime-dependent drift and diffusion terms, the switching components, and the endogeneity in regime changes. While we find strong evidence of regime-switching effects, models with a switching diffusion term capture the VIX dynamics considerably better than models with only a switching drift, confirming the presence and importance of volatility regimes. Strong evidence of nonlinear endogeneity in regime changes is also detected. Meanwhile, we find significant nonlinearity in the regime-dependent diffusion specification, suggesting that the nonlinearity in the VIX dynamics cannot be accounted for by regime-switching effects alone. Finally, we find that models based on the CIRCEV specification are significantly closer to the true data generating process of VIX than models based on the NLDCEV specification uniformly across all regime-switching specifications.

Accurately measuring and forecasting value-at-risk (VaR) remains a challenging task at the heart of financial economic theory. Recently, quantile regression models have been used successfully to capture the conditional quantiles of returns and to forecast VaR accurately. In this paper, we further explore nonlinearities in data and propose to couple realized measures with the nonlinear quantile regression framework to explain and forecast the conditional quantiles of financial returns. The nonlinear quantile regression models are implied by the copula specifications and allow us to capture possible nonlinearities, tail dependence, and asymmetries in the conditional quantiles of financial returns. Using high frequency data that covers most liquid US stocks in seven sectors, we provide ample evidence of asymmetric conditional dependence with different levels of dependence, which are characteristic for each industry. The backtesting results of estimated VaR favour our approach.

This paper presents a no-arbitrage yield-curve model that explicitly incorporates the central-bank policy rate. This model is consistent with the existence of a lower bound for nominal interest rates, which makes it particularly relevant in the current context of extremely low interest rates. Changes in the policy rates depend on the monetary-policy phase, that can be either in an easing, status quo or tightening mode. The estimation of the model, based on daily euro-area yield data, reveals the strong influence of the monetary-policy phases on the shape of the yield curve. This relationship can, in turn, be exploited to estimate the probabilities of being in the different monetary-policy phases. The model is also used to compute term premiums, that are the parts of the yields reflecting the aversion of investors to interest rate risk. The results point to the existence of statistically significant premiums for many dates, even for short horizons.