A new frontier for studying within-person variability: Bayesian multivariate generalized autoregressive conditional heteroskedasticity models.

PurposeThis paper aims to test whether the latest global financial crisis propagated contagiously from the USA to the rest of the world.Design/methodology/approachIf the reason of the propagation of a crisis is a normal time interdependence with the crisis origin country due to real linkages, the spread of crisis can be limited by implementing well-defined preventive policies. On the other hand, if a crisis propagates because of the speculative attacks or irrational behaviors, the “national policymakers will face difficulties in protecting their markets from such a crisis” (Kleimeier et al., 2003, p. 2). Therefore, separation of contagion and interdependence may provide crucial insights for policymakers to implement appropriate policies to prevent and/or stop the financial crisis. Hence, this paper compares the heteroscedasticity-corrected conditional correlations and dynamic conditional correlations in the tranquil and shock periods.FindingsThe findings were quite straightforward and consistent for both Forbes and Rigobon heteroscedasticity correction technique and dynamic conditional correlation (DCC) model. The Forbes and Rigobon technique failed to reject the null of no contagion for 25 countries in our data sample, while the DCC model failed to reject the null of no contagion for 21 countries. While heteroscedasticity-corrected correlation technique confirmed the presence of a contagion for six countries, the DCC technique confirmed the presence of a contagion for ten countries.Originality/valueThis study particularly investigates whether the subprime mortgage crisis spilled over contagiously to the rest of the world. To investigate whether there is a significant increase in the cross-market correlations between the crisis origin country, the USA and the rest of the world markets during the latest financial crisis, both heteroscedasticity-corrected correlation technique and DCC model are used.Therefore, this study possibly contributes well to the literature using a large country set and conducting the analysis from different angles for important properties.

금융 시계열 자료들 간의 상관계수는 자산의 배분, 위험관리 그리고 포트폴리오의 선택에 있어서 중요한 역할을 한다. 이러한 상관계수들을 모형화하기 위해 단변량-GARCH 모형을 다변량-GARCH 모형으로 확장시킨 MGARCH류 모형들에 대한 많은 연구들이 진행되고 있다. 특히, CCC 모형 (Bollerslev, 1990)과 DCC 모형 (Engle, 2002)은 다른 모형들에 비해 추정해야 할 모수의 수가 작다는 이점으로 인해 분석에 널리 쓰이고 있다. 본 논문에서는 국내 주가자료에 대해 CCC 모형과 DCC 모형을 적합시킨 후, 각 모형들에 대한 VaR(value at risk)와 사후검증(back-testing), 결합예측영역(joint prediction region) 등을 통하여 두 모형의 예측 능력을 비교해 보고자 한다. Conditional correlation between financial time series plays an important role in risk management, asset allocation and portfolio selection and therefore diverse efforts for modeling conditional correlations in multivariate-GARCH processes have been made in last two decades. In particular, CCC (cf. Bollerslev, 1990) and DCC(dynamic conditional correlation, cf. Engle, 2002) models have been commonly used since they are relatively parsimonious in the number of parameters involved. This article is concerned with DCC modeling for multivariate GARCH processes in comparison with CCC specification. Various multivariate financial time series are analysed to illustrate possible advantages of DCC over CCC modeling.

Regularized semiparametric estimation of high dimensional dynamic conditional covariance matrices

A Model of Price Correlations between Clean Energy Indices and Energy Commodities

The paper develops two Dynamic Conditional Correlation (DCC) models, namely the Wishart DCC (WDCC) model and the Matrix-Exponential Conditional Correlation (MECC) model. The paper applies the WDCC approach to the exponential GARCH (EGARCH) and GJR models to propose asymmetric DCC models. We use the standardized multivariate t-distribution to accommodate heavy-tailed errors. The paper presents an empirical example using the trivariate data of the Nikkei 225, Hang Seng and Straits Times Indices for estimating and forecasting the WDCC-EGARCH and WDCC-GJR models, and compares the performance with the asymmetric BEKK model. The empirical results show that AIC and BIC favour the WDCC-EGARCH model to the WDCC-GJR and asymmetric BEKK models. Moreover, the empirical results indicate that the WDCC-EGARCH-t model produces reasonable VaR threshold forecasts, which are very close to the nominal 1% to 3% values.

A Bayesian conditional autoregressive geometric process model for range data

Complexity analysis based on generalized deviation for financial markets

Estimation and empirical performance of non-scalar dynamic conditional correlation models

We test for the contagion effects stemming from the Greek debt crisis in the daily 10-year sovereign bond yield spreads in nine Economic and Monetary Union (EMU) countries. To this end, we estimate the dynamic conditional correlation (DCC) model of Engle (2002) from January 01, 2003 to December 31, 2015. In addition, we calculate and plot the upper and lower bounds of the confidence interval for each DCC series. To the best of our knowledge, this approach of Kchaou and Bellalah (2020) has never been used to study the contagion of the subprime and Greek crises between the 10-year sovereign bond yield spreads of the main EMU countries. Consequently, this approach enables us to compare our results with those of previous works based on other methods. It also offers useful insights to policy makers to address the contagion effect through the implementation of adequate measures. Although the Greek spread played the role of a global factor for the majority of countries during the observation period, the results invalidate the existence of contagious episodes resulting from the Hellenic crisis. We justify these findings either by the weakness of the weight of the Greek economy in the euro area or by the effectiveness of the unconventional monetary policies taken by the European Central Bank (ECB), the bailouts for Greece in 2010, 2012 and 2015 and the austerity measures and structural reforms implemented by the governments of EMU countries. Moreover, DCC between Greece and the other countries have shown a downward behavior during the acute phases of Greek crisis, suggesting a disintegration of the Hellenic bond market with those of other euro area countries during periods of financial turmoil. Furthermore, the results indicate that the subprime crisis affected a large part of these markets well before the bankruptcy of Lehman Brothers. All these findings provide valuable information for international investors, central bankers and policymakers.

SummaryWe propose a flexible generalized auto-regressive conditional heteroscedasticity type of model for the prediction of volatility in financial time series. The approach relies on the idea of using multivariate B-splines of lagged observations and volatilities. Estimation of such a B-spline basis expansion is constructed within the likelihood framework for non-Gaussian observations. As the dimension of the B-spline basis is large, i.e. many parameters, we use regularized and sparse model fitting with a boosting algorithm. Our method is computationally attractive and feasible for large dimensions. We demonstrate its strong predictive potential for financial volatility on simulated and real data, and also in comparison with other approaches, and we present some supporting asymptotic arguments.

An important development in the modelling of exchange rate volatility is the use of artificial neural networks (ANN) to create enhanced generalised autoregressive conditional heteroskedasticity (GARCH) models. Conventional GARCH models are good at capturing the clustering of volatility in financial time series, but they have trouble understanding complex linkages and non-linear patterns in the data. This paper aims to investigate the hybrid approach in modelling exchange rate volatility of two currency pairs: South African Rand against Brazilian Real (ZAR/REAL) and South African Rand against Chinese Yuan (ZAR/YUAN) using monthly observations over the period of January 1996-March 2024. The paper introduced ANN as an additional factor to both symmetric and asymmetric GARCH models that capture most common stylised facts about exchange rate volatility and leverage effects. The GARCH (1,1)-ANN, EGARCH (1,1)-ANN, and GJR-GARCH (1,1)-ANN models were used, and their performance was assessed using mean squared error (MSE), mean absolute error (MAE), and mean absolute percentage error (MAPE). The results revealed that the EGARCH (1,1)-ANN model had the best overall performance when compared to the other hybrid models based on all three evaluation measures for both the currency pairs’ data. The paper recommends further similar studies to predict future exchange rate trends and also incorporating other nonlinear methods.

Realized volatility or price range: Evidence from a discrete simulation of the continuous time diffusion process

The modeling of high-frequency volatility is of utmost importance in comprehending market dynamics and the characteristics of risk. The adoption of high-frequency volatility modeling in the agricultural sector has the potential to enhance risk management for food production enterprises through the utilization of hedging strategies to mitigate the impact of food price changes. The GARCH (Generalized Autoregressive Conditional Heteroskedasticity) class of models is well-suited for the analysis and prediction of volatility in financial time series. These models effectively capture the highest level of volatility observed in the time series data. Therefore, this research utilizes one-minute trade data of the Wind Agriculture Index (886045.WI) to examine the efficacy and predictability of several GARCH class volatility models. When it comes to fitting the data within the sample, the TGARCH model demonstrates superior performance compared to both the GARCH and EGARCH models. The fluctuations in the price changes of the Wind Agriculture Index exhibit characteristics of time variability and clustering, which can be attributed to the relatively low barriers for entry and exit in the agricultural planting sector. Simultaneously, within the agricultural market, an imbalance is observed whereby the influence of positive information on price volatility surpasses that of negative information.

Summary We propose a new generalized autoregressive conditional heteroscedastic (GARCH) model with tree-structured multiple thresholds for the estimation of volatility in financial time series. The approach relies on the idea of a binary tree where every terminal node parameterizes a (local) GARCH model for a partition cell of the predictor space. The fitting of such trees is constructed within the likelihood framework for non-Gaussian observations: it is very different from the well-known regression tree procedure which is based on residual sums of squares. Our strategy includes the classical GARCH model as a special case and allows us to increase model complexity in a systematic and flexible way. We derive a consistency result and conclude from simulation and real data analysis that the new method has better predictive potential than other approaches.

The major topic investigates how classical methods (ARCH and GARCH) and well-known machine learning algorithms, support vector regression, and hybrid methods. This paper aims to predict and forecast volatility to develop a two-stage forecasting approach the volatility of the Amman Stock Exchange Index (ASE) effectively. Additionally, the effectiveness of the machine learning techniques’ selection and utilization of information in stock data is evaluated. Methods the semiparametric estimating technique known as support vector regression (SVR) has been widely used for the prediction of volatility in financial time series. By integrating SVR with the GARCH model (GARCH-SVR) application with various kernels (Radial Basis Kernel Function (RBF), Polynomial Kernel Function (PF), and linear Kernel Function (lF)). The suggested learning approaches are compared to two well-known statistical time series models, Autoregressive Conditional Heteroskedasticity (ARCH) and Generalized Autoregressive Conditional Heteroskedasticity (GARCH), in order to assess the assertion that they can properly anticipate ASE volatility. To compare their results, RMSE is employed as an error metric. There is evidence that the GARCH-SVR model performs best for predicting volatility time series, and classical volatility model techniques have an enormous predictive performance better than machine learning models.