ALBERT

Description: This paper investigates dependence among insurance claims arising from different lines of business (LoBs). Using bivariate and multivariate portfolios of losses from different LoBs, we analyse the ability of various copulas in conjunction with skewed generalised hyperbolic (GH) marginals to capture the dependence structure between individual insurance risks forming an aggregate risk of the loss portfolio. The general form skewed GH distribution is shown to provide the best fit to univariate loss data. When modelling dependency between LoBs using one-parameter and mixture copula models, we favour models that are capable of generating upper tail dependence, that is, when several LoBs have a strong tendency to exhibit extreme losses simultaneously. We compare the selected models in their ability to quantify risks of multivariate portfolios. By performing an extensive investigation of the in- and out-of-sample Value-at-Risk (VaR) forecasts by analysing VaR exceptions (i.e. observations of realised portfolio value that are greater than the estimated VaR), we demonstrate that the selected models allow to reliably quantify portfolio risk. Our results provide valuable insights with regards to the nature of dependence and fulfils one of the primary objectives of the general insurance providers aiming at assessing total risk of an aggregate portfolio of losses when LoBs are correlated.

Description: This paper illustrates how outliers can affect both the estimation and testing of leverage effect by focusing on the TGARCH model. Three estimation methods are compared through Monte Carlo experiments: Gaussian Quasi-Maximum Likelihood, Quasi-Maximum Likelihood based on the Student-t likelihood and Least Absolute Deviation method. The empirical behavior of the t-ratio and the Likelihood Ratio tests for the significance of the leverage parameter is also analyzed. Our results put forward the unreliability of Gaussian Quasi-Maximum Likelihood methods in the presence of outliers. In particular, we show that one isolated outlier could hide true leverage effect whereas two consecutive outliers bias the estimated leverage coefficient in a direction that crucially depends on the sign of the first outlier and could lead to wrongly reject the null of no leverage effect or to estimate asymmetries of the wrong sign. By contrast, we highlight the good performance of the robust estimators in the presence of one isolated outlier. However, when there are patches of outliers, our findings suggest that the sizes and powers of the tests as well as the estimated parameters based on robust methods may still be distorted in some cases. We illustrate these results with two series of daily returns.

Description: In dealing with a panel of seasonal data with cross-section dependence, this paper establishes a common factor model to investigate whether the seasonal and non-seasonal non-stationarity in a series is pervasive, or specific, or both. Without knowing a priori whether the data are seasonal stationary or not, we propose a procedure for consistently estimating the model; thus, the seasonal non-stationarity of common factors and idiosyncratic errors can be separately detected accordingly. We evaluate the methodology in a series of Monte Carlo simulations and apply it to test for non-stationarity and to disentangle their sources in panels of worldwide real exchange rates and of consumer price indexes for 37 advanced economies.

Description: Several studies have established the predictive power of the yield curve i.e. the difference between long and short-term bond rates and the role of asymmetries in the term structure of bond yields with respect to real economic activity. Using an extensive dataset, comprising 3-month, 1-year, 5-year and 10-year constant maturity Treasury bonds for the Eurozone southern periphery countries – the so-called “PIIGS” – from January 1999 to April 2019, we investigate the links between bond yields of different maturities for the Eurozone southern peripheral countries and we find they co-evolve in line with the predictions of the Expectations Hypothesis theory. We demonstrate the presence of nonlinearities in the term structure, and utilize a multivariate asymmetric two-regime Markov-switching VAR methodology to model them properly. Moreover, we address the economic reasoning behind the introduction of an equilibrium-correction regime-switching approach, hence providing potentially important insights on the behaviour of the entire yield curve. We reveal that the regime shifts are related to the state of the business cycle, particularly in economies in which monetary policy decisions are implemented via changes in short-term rates as a response to deviations of output from equilibrium levels. Our results may have important statistical and economic implications on the behaviour of the term structure of bond yields.

Description: This paper investigates how the Federal Reserve (Fed) and the Bank of England, Bank of Japan and the European Central Bank reacted in the aftermath of the financial crisis by making use of both conditional and unconditional interest rate quantiles regressions and data on shadow short rate of interest and a measure of uncertainty. Firstly, the unconditional quantile regression offers some support for increased reaction by the Fed as the ZLB is approached. Secondly, the decreased reaction of the Fed and other monetary policy makers towards uncertainty particularly at lower conditional quantiles of interest rates lends support to expansionary mechanism in place during this time. Hence uncertainty is key to policy reaction, and more so during episodes of crisis.

Description: We compare local and global polynomial solution methods for DSGE models with Epstein- Zin-Weil utility. We show that model implications for macroeconomic quantities are relatively invariant to choice of solution method but that a global method can yield substantial improvements for asset prices and welfare costs. The divergence in solution quality is highly dependent on parameters which affect value function sensitivity to TFP volatility, as well as the magnitude of TFP volatility itself. This problem is pronounced for calibrations at the extreme of those accepted in the asset pricing literature and disappears for more traditional macroeconomic parameterizations.

Description: Howell Tong (simplified Chinese: 汤家豪) is a pioneer in the field of nonlinear time series analysis, linking it with deterministic chaos. He is the father of the threshold time series models, which have extensive applications in ecology, economics, epidemiology and finance. Since October 1, 2009, he has been an Emeritus Professor at the London School of Economics and was twice (2009, 2010) holder of the Saw Swee Hock Professorship of Statistics at the National University of Singapore. He was a Distinguished Visiting Professor of Statistics at the University of Hong Kong from 2005 to 2013. He got a Master in Science degree in 1969 and a Doctor of Philosophy degree in 1972 from the University of Manchester Institute of Science and Technology (UMIST), where he studied under Maurice Priestley. From 1999 to September 2009, Tong was Chair of Statistics at LSE and founded the Centre for the Analysis of Time Series. Between 1997 and 2004, Tong was also Chair Professor of Statistics, Founding Dean of the Graduate School and later Pro-Vice Chancellor, University of Hong Kong. He was elected a Fellow of the Institute of Mathematical Statistics in 1993, an Honorary Fellow of the Institute of Actuaries, England in 1999, and a Foreign Member of the Norwegian Academy of Science and Letters in 2000. In 2000, he became the first statistician to win the (class II) the State Prize in Natural Sciences in China. In 2002, the University of Hong Kong gave him their then-highest award, the Distinguished Research Achievement Award, carrying with it a research grant of HK$1,000,000 per annum for three years. The Royal Statistical Society, UK, awarded him their Guy Medal in Silver in 2007 in recognition of his “…many important contributions to time series analysis over a distinguished career and in particular for his fundamental and highly influential paper ‘Threshold autoregression, limit cycles and cyclical data,’ read to the Society in 1980, which paved the way for a major body of work in non-linear time series modelling.” In 2012, the International Chinese Statistical Association awarded him the Distinguished Achievement Award. Tong is also a Distinguished Professor-at-Large at the University of Electronic Science and Technology of China in Chengdu, China, and a Distinguished Visiting Professor at Tsinghua University in Beijing, China.

Description: In light of several economic and financial crises and institutional changes experienced by the European countries, we examine whether these economies achieved synchronization of their business cycles and fostered synchronization of their growth rates. Controlling for reverse causality, we conduct multiple endogenous break tests and find that (i) several endogenous break dates correspond to idiosyncratic shocks affecting individual countries or major shocks in international arena but not the adoption of the euro; this result suggests that the convergence process has been nonlinear for a number of countries and that studies imposing break dates exogenously, such as the launch of euro, may lead to biased conclusions; (ii) while output growth was increasingly synchronized for some countries, integration occurred in an asymmetric way and it did not change or did not occur for others despite being in the same common currency area (iii) convergence has been prevalent among the non-Eurozone economies in our sample.

Description: The rise in US partisan conflict following the Great Recession led to a popular belief that uncertainty about fiscal policy was impeding output growth. I explore this hypothesis by nesting it in a standard structural vector autoregression (SVAR) model traditionally used for estimating fiscal multipliers. I augment the model with stochastic volatility (a measure of uncertainty) and allow that to interact with the endogenous variables. I consider various trend assumptions, subsamples and information sets and find that the evidence does not support this hypothesis. The results reveal that there is no systematic relationship between fiscal policy uncertainty and output. Moreover, a time-varying parameter version of the model shows that the lack of consistency across specifications is not driven by changes in the transmission of uncertainty shocks over time. Finally, I revisit Fernández-Villaverde, Guerrón-Quintana, Kuester, and Rubio-Ramírez (Fernández-Villaverde, J., P. Guerrón-Quintana, K. Kuester, and J. Rubio-Ramírez. 2015. “Fiscal Volatility Shocks and Economic Activity.” American Economic Review 105: 3352–3384) who find a significant negative relationship between fiscal policy uncertainty and output. I show that when their estimation is modified to incorporate the uncertainty around the estimate of uncertainty, their empirical result falls in line with the findings in this paper.

Description: This paper shows how stock market volatility regimes affect the cross-section of stock returns along quality and liquidity dimensions. We find that, during crisis periods, low quality and low liquidity stocks experience relatively higher losses than predicted in normal times, while high quality and high liquidity stocks experience rather relatively lower losses. These findings lend strong support to the presence of cross-market and within-market flight-to-quality and to-liquidity episodes during crisis periods. During low volatility periods, however, low quality and low liquidity stocks earn relatively larger returns, while high quality and high liquidity stocks yield lower returns; suggesting that low volatility conditions benefit junk and illiquid stocks but not quality and liquid stocks. Finally, our results reveal that liquidity level dominates liquidity beta in explaining stock returns across the different market volatility regimes.