Dynamic Asset Correlations Based on Vines

Regularized semiparametric estimation of high dimensional dynamic conditional covariance matrices

The number of empirical studies using daily stock returns is rapidly increasing.' Researchers generally assume that the distribution of stock returns is identical for all days of the week-a convenient statistical assumption but not a necessary condition of market equilibrium. Nevertheless, there are reasons to suspect that the distribution of returns may vary according to the day of the week; the most obvious is the impact of weekends on Monday's return. Since Monday's return is calculated over three instead of one calender day, the mean and variance may well be higher on Monday compared with any other daily return (perhaps three times as large). Previous research has examined the daily distribution of stock returns for a Monday effect. While Fama (1965) does not compare daily mean returns, he does report that Monday's variance is about 20o greater than other daily returns. With A traditional distributional assumption regarding the returns on a financial asset specifies that the expected returns are identical for all days of the week. Contrary to this plausible assumption, this paper discovers that the expected returns on common stocks and treasury bills are not constant across days of the week. The most notable evidence is for Monday's returns where the mean is unusually low or even negative. Several explanations of the results are investigated, but none proves satisfactory. Aside from documenting significant day of the week effects, the implications of the results for tests of market efficiency are examined. While market-adjusted returns continue to exhibit day of the week effects, these effects are no longer concentrated on Monday. * During the last year we have had many conversations with our colleagues throughout the professions. We are particularly thankful to the participants at sessions of the Western and American Finance Associations where preliminary versions of this paper were presented. Special thanks are due to Gene Fama, Gailen Hite, Jon Ingersoll, Robert Jarrow, Merton Miller, Clifford Smith and Rex Thompson. The expert research assistance of Frank Rabinovitch and YoungSup Yun is gratefully acknowledged. 1. The list of papers using daily stock returns is too long to name. Recent examples include Charest (1978), Dodd (1980), Banz (1981), and Reinganum (1981).

The uncertainty of conditional returns, volatilities and correlations in DCC models

By analysing the Dynamic Conditional Correlations (DCC) of the daily stock returns of four OECD countries with that of the US for the period 2006–2010, we could find a process of increasing correlations (contagion) in the first phase of the US financial crisis and an additional increase of correlations (herding) during the second phase of the US financial crisis for the UK, Australia and Switzerland. However, the impact of the US financial crisis on Japan was limited to the increase in correlation volatilities in the first phase. We also propose a new approach (DCC Multivariate Generalized Autoregressive Conditional Heteroscedastic model with Exogenous variables (DCCX-MGARCH)) that allows simultaneous estimation of the DCC and their determinants, which can be used to identify channels of contagion. It is shown that an increase in VIX stock market index increases conditional correlations but an increase in the TED spread and relative stock market capitalization decrease conditional correlations of stock returns between four OECD countries and the US.

We study the properties of the quasi-maximum likelihood estimator (QMLE) and related test statistics in dynamic models that jointly parameterize conditional means and conditional covariances, when a normal log-likelihood os maximized but the assumption of normality is violated. Because the score of the normal log-likelihood has the martingale difference property when the forst two conditional moments are correctly specified, the QMLE is generally Consistent and has a limiting normal destribution. We provide easily computable formulas for asymptotic standard errors that are valid under nonnormality. Further, we show how robust LM tests for the adequacy of the jointly parameterized mean and variance can be computed from simple auxiliary regressions. An appealing feature of these robyst inference procedures is that only first derivatives of the conditional mean and variance functions are needed. A monte Carlo study indicates that the asymptotic results carry over to finite samples. Estimation of several AR and AR-GARCH time series models reveals that in most sotuations the robust test statistics compare favorably to the two standard (nonrobust) formulations of the Wald and IM tests. Also, for the GARCH models and the sample sizes analyzed here, the bias in the QMLE appears to be relatively small. An empirical application to stock return volatility illustrates the potential imprtance of computing robust statistics in practice.

PurposeThis study aims to investigate the possibilities of cryptocurrencies as hedges and diversifiers in the Indian stock market before and during financial crisis due to the pandemic and the Russia–Ukraine war.Design/methodology/approachResearchers have used daily data on cryptocurrencies and Indian stock prices from March 10, 2015 to August 26, 2022. The researchers have used the dynamic conditional correlations (DCC)-GARCH model to determine the volatility spillover and dynamic correlation between stocks and digital currencies. Further, researchers have explored hedge ratio, portfolio weight and hedging effectiveness using the estimates of the DCC-GARCH model.FindingsThe findings indicate a negative conditional correlation between equities and cryptocurrencies before the crisis and a positive conditional correlation except for Tether during the crisis. Which implies that cryptocurrencies serve as a hedging asset in the stock market before a crisis but are not more than a diversifier during the crisis, except for Tether. Notably, Tether serves as a safe haven during times of crisis. Finally, the study suggests that Bitcoin, Ethereum, Binance Coin and Ripple are the most effective diversifiers for Indian stocks during the crisis.Originality/valueThis study makes several contributions to the existing literature. First, it compares the hedge and diversification roles of cryptocurrencies in the Indian stock market before and during crisis. Second, the study findings provide insights on risk hedging and can serve as a guide for investors. Third, it may help rational investors avoid underestimating risk while constructing portfolios, particularly in times of financial turmoil.

PurposeThe purpose of this paper is to investigate the time series behavior of co‐movements among 11 European real estate securities markets, with each other as well as between country‐averages, over the sample period from January 1999 to January 2010 by utilizing the asymmetric dynamic conditional correlation (ADCC) technique, long‐memory tests and multiple structural break methodology.Design/methodology/approachFirst the ADCC from the multivariate GJR‐GARCH model is used to estimate the pair‐wise conditional correlations between the 11 securitized real estate markets. Then, the 11 country‐average conditional correlation series is subject to a battery of four long‐memory tests to form an “on the balance of evidence” picture; the semi‐parametric Geweke and Porter‐Hudak procedure and Robinson test, as well as the non‐parametric Hurst‐Mandelbrot R/S and Lo's modified R/S tests. Finally, the Bai and Perron's multiple structural break methodology seeks to test whether the average conditional correlations are subject to regime switching via the detection of breaks in the co‐movements of real estate securities returns.FindingsLow to moderate conditional correlations are found for these European real estate securities market and a higher level of correlation in the aftermath of the global financial crisis. The long‐memory correlation effect is present for nine European real estate securities markets. In addition, the conditional correlations are subject to regime switching with two structural breaks in four country‐average correlation series. Across the regimes, a higher level of correlation is linked to a higher level of volatility and a lower level of return, and this happened around the global financial crisis period.Research limitations/implicationsThe findings that national real estate securities correlations exhibit time‐varying and asymmetric behavior can help investors understand how real estate securities will co‐move in different market scenarios (e.g. “crisis” and “non‐crisis” times). Moreover, the process of dynamic covariance analysis and forecasting (the ultimate objective in portfolio management) should not rely too much on short‐term autoregressive moving average models. Instead, a combination of some appropriate long‐range dependence models and regime‐switching specifications is needed.Originality/valueThis paper offers useful insights into the time series behavior of average dynamic conditional correlations in European public property markets.

The Intertemporal Relation between Risk and Returns in Australia

Dynamic correlation analysis of financial contagion: Evidence from the Central and Eastern European markets

The purpose of this paper is to investigate dynamic changes of the volatilities among stock indices, exchange rates, and interest rates in Taiwan, America, and South Korea during and after the period of American 9/11 terrorist attacks. Using sample data from October 1, 1999 to January 24, 2006, this study applies the Dynamic Conditional Correlation Multivariate GARCH model proposed by Engle (2002) to estimate the volatilities of markets themselves and the dynamic conditional correlations between different markets. The empirical results show: (1) during the period of 9/11 incident, both the short-run volatilities of every market and the short-run dynamic correlations between different markets enlarged significantly, but all correlations with a time lag of one day; (2) after 9/11 incident, the long-run volatilities of stock and exchange markets in all countries enlarged significantly, but the volatilities of interest rate markets diminished significantly or changed insignificantly; the long-run dynamic correlations between the same markets of different countries changed significantly, but the correlations between different markets of the same countries changed insignificantly.

The research article presents an overall comparative study of multivariate GARCH M-GARCH in portfolio risk management, where three prevailing specifications VEC, CCC and DCC models are considered. We take daily closing prices of four major assets of the year 2018 through to 2023; S&P 500, NASDAQ-100, gold futures, and US Treasury Bonds, to estimate conditional variances, covariances, and dynamic correlations using maximum likelihood estimation. The descriptive statistics indicate that volatility is highly concentrated in clustering and time varying across assets with NASDAQ having the highest volatility (2.08) and significant negative skewness that shows non-normal returns. Comparison on models based on information criteria indicates that the Dynamic Conditional Correlation (DCC) specification has better performance with less computation need, fewer 8 parameters as compared to 21 (VEC) and greater log-likelihood with high improvement (164.67 units). Adequate model specification is shown through diagnostic testing using Ljung-Box test and ARCH-LM tests. Empirical results indicate that the volatility persistence (alpha + - 0 = 0.98) and the dynamics of high correlation (beta = 0.9321) are high which indicates long-memory properties and mean-reverting behavior which does not support constant correlation assumptions.

Exogenous shocks can affect equity markets by changing volatility and cross-market co-movement. This study examines how two U.S.-centred events, treated as different shock types, influence time-varying conditional correlations between the U.S. stock market and other G7 markets. The uncertainty shock is proxied by the U.S. presidential election of 5 November 2024, while the action shock is proxied by President Trump’s 2 April 2025 announcement of reciprocal tariffs. Using daily log returns for the S&P 500 and leading indices for Canada, France, Germany, Italy, Japan and the United Kingdom, we cover January 2010 to July 2025 and assess event effects using correlation paths for June 2024–June 2025 and symmetric ±30-day windows. We employ a DCC-GARCH model to jointly estimate conditional variances and dynamic correlations for six USA-G7 pairs. The results indicate persistent correlation dynamics, with Canada/USA the highest and Japan/USA the lowest. Election-related uncertainty is associated with declines in correlation for European pairs, suggesting temporary decoupling, while Canada and Japan show only small changes. By contrast, the tariff action shock significantly increases conditional correlations across all country/USA pairs, implying stronger market synchronisation, with the largest increases in North America and parts of Europe, and the smallest adjustment in Japan.

The partial correlation is a commonly used measure for assessing the bivariate correlation of two quantitative variables after eliminating the influence of one or more other variables. The partial correlation is generally interpreted as the correlation that would result if the variables to be eliminated were fixed (not allowed to vary and influence the other variables), which is referred to in the statistical literature as conditional correlation. The present paper demonstrates, by means of theoretical derivations and practical examples, that when the assumption of multivariate normality is violated (e.g., as a result of nonlinear relationships among the variables investigated) the usual interpretation of the partial correlation coefficient will be basically incorrect. In extreme cases the value of the partial correlation coefficient may be strongly positive, close to 1, whereas the conditional correlation may have a large negative value. To solve this problem the paper suggests to partial out a certain function (in most cases the square) of the variables whose effects are to be eliminated if nonlinear relationships are likely to occur.

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.

Asymmetric connectedness between cryptocurrency environment attention index and green assets