Center of Risk Management

HEC Lausanne

Université de Lausanne

1015 Lausanne

Thibault Vatter, Hau-Tieng Wu, Valérie Chavez-Demoulin, and Bin Yu

Non-parametric estimation of intraday spot volatility: disentangling instantaneous trend and seasonality

November - 2013


We provide a new framework to model trends and periodic patterns in high-frequency financial data. Seeking adaptivity to ever changing market conditions, we enlarge the popular Fourier flexible form into a richer (time- varying) functional class and relax the assumptions from usual (static) intraday models in two directions. First, the realized volatility (i.e. the slowly time-varying component) is modelled as an instantaneous trend which evolves in real time. Second, the seasonality is no longer assumed constant over the sample. We provide the estimators associated with our class models and show that they have low variance and are essentially un- biased. As an application, we analyze the trajectories of the spot volatility in the foreign exchange market. Our results suggest that failing to factor in the seasonality’s dynamic properties may lead to severe underestima- tion/overestimation of the intraday spot volatility.

To download the paper click here



Valérie Chavez-Demoulin, Paul Embrechts, Marius Hofert

An Extreme Value Approach For Modeling Operational Risk Losses Depending on Covariates

Updated November - 2013


A general methodology for modeling loss data depending on covariates is developed. The parameters of the frequency and severity distributions of the losses may depend on covariates. The loss frequency over time is modeled via a non-homogeneous Poisson process with integrated rate function depending on the covariates. This corresponds to a generalized additive model which can be estimated with spline smoothing via penalized maximum likelihood estimation. The loss severity over time is modeled via a nonstationary generalized Pareto model depending on the covariates. Whereas spline smoothing can not be directly applied in this case, an efficient algorithm based on orthogonal parameters is suggested. The methodology is applied to a database of (mostly banking) operational risk losses. Estimates, including confidence intervals, for risk measures such as Value-at-Risk or Expected-Shortfall as required by the Basel II/III framework are computed. We provide links to a detailed R implementation of the statistical methodology.

To download the paper click here



Robert Engle, Eric Jondeau, and Michael Rockinger

Dynamic Conditional Beta and Systemic Risk in Europe

March - 2012


Systemic risk can be defined as the propensity of a financial institution to be under-capitalized when the financial system as a whole is under-capitalized. It combines the market capitalization of the firm, the sensitivity of its equity return to market shocks, and its financial leverage. In this paper, we describe an econometric approach designed to measure systemic risk for non-U.S. institutions. We extend the approach developed by Brownlees and Engle (2010) to the case with several factors explaining the dynamic of financial firms' return and with asynchronicity of the time zones. Our model combines a DCC model to estimate the dynamic of the beta parameters, univariate GARCH models to estimate the dynamic of the volatility of the error terms, and a dynamic t copula to estimate the dynamic of the dependence structure between the innovations. We apply this methodology to the 194 largest European financial firms and estimate their systemic risk over the 2000-2012 period. We find that banks and insurance companies bear about 80% and 20% of the systemic risk in Europe, whereas systemic risk is essentially unaffected by financial services and real estate firms. Over the recent period, the systemically riskiest countries are the UK and France, and the riskiest firms are Deutsche Bank and BNP Paribas.

To download the paper click here



Ari-Pekka Hameri and Johua Hintsa

Assessing the Drivers of Change for Cross-Border Supply Chains



International Journal of Physical Distribution & Logistics, 39(9), pp.741-761.



Purpose – This paper aims to systematically document drivers of change and the implications they will have on international supply chain management in the coming two decades.

Design/methodology/approach – This study was commissioned by the World Customs Organization (WCO) at the end of June 2006. Because of increased trade volumes, emerging complex supply networks and heightened security concerns, the WCO saw the need to assess future trends and drivers in supply chain management. The Delphi method was applied to identify a set of foreseeable drivers of change and to assess their predicted impact on global supply chain management in the coming ten to 20 years. Based on a literature review of 150 recent publications and interviews among 33 industry, academic and customs experts, a survey was designed and conducted to collect current and potential change drivers in global supply chains. These drivers were compiled and prioritized by an eclectic team of 12 specialists.

Findings – The main results of the study are strongly connected to strategic and operational supply chain planning for the next ten to 20 years. They are related to increased off-shoring of operations through truly global manufacturing, characterized by its intercontinental supply of materials; increased product complexity with shorter product life cycles; increased importance of business-to-government networking for operational and security efficiency; introduction of new supply chain services integrating financial, physical and information flows leading to further consolidation in the logistics markets; and the overall increase in risks and vulnerabilities in international supply chains.

Originality/value – This paper provides a 360 degree view of the future of international supply chain management and the challenges companies will face to compete in the twenty-first century business environment.