33 publications classées par:
type de publication
: Revue avec comité de lecture
Articles Cai J.-J. Chavez-Demoulin V. & Guillou A. (soumis à l'éditeur). Estimation of the Marginal Expected Shortfall in the Context of an Infinite Mean Model. http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2742090.
Chavez-Demoulin V. Embrechts P. Hofert M. (in press). An extreme value approach for modeling Operational Risk losses depending on covariates. Journal of Risk and Insurance.
Niemi T. Appelqvist P. Babongo F. Chavez-, Demoulin V. & Hameri A.-P. (in press). Weather and supply chain performance in sport goods distribution. International Journal of Retail & Distribution Management.
Babongo Bosombo F., Chavez-Demoulin V., Hameri A.-P. & Niemi T. (2016). Weather and supply chain performance in sport goods distribution. International Journal of Retail & Distribution Management, 44(2), 178 - 202. [doi] [abstract]
Purpose¦- The purpose of this paper is to study how variations in weather affect demand and supply chain performance in sport goods. The study includes several brands differing in supply chain structure, product variety and seasonality.¦Design/methodology/approach¦- Longitudinal data on supply chain transactions and customer weather conditions are analysed. The underlying hypothesis is that changes in weather affect demand, which in turn impacts supply chain performance.¦Findings¦- In general, an increase in temperature in winter and spring decreases order volumes in resorts, while for larger customers in urban locations order volumes increase. Further, an increase in volumes of non-seasonal products reduces delays in deliveries, but for seasonal products the effect is opposite. In all, weather affects demand, lower volumes do not generally improve supply chain performance, but larger volumes can make it worse. The analysis shows that the dependence structure between demand and delay is time varying and is affected by weather conditions.¦Research limitations/implications¦- The study concerns one country and leisure goods, which can limit its generalizability.¦Practical/implications¦- Well-managed supply chains should prepare for demand fluctuations caused by weather changes. Weekly weather forecasts could be used when planning operations for product families to improve supply chain performance.¦Originality/value¦- The study focuses on supply chain vulnerability in normal weather conditions while most of the existing research studies major events or catastrophes. The results open new opportunities for supply chain managers to reduce weather dependence and improve profitability.
Abaunza F., Chavez-Demoulin V., Hameri A.-P. & Niemi T. (2015). Do flow principles of operations management apply to computing centres?. Production Planning & Control: The Management of Operations, 26(4), 249-264. [doi] [url] [web of science] [abstract]
By analysing large data-sets on jobs processed in major computing centres, we study how operations management principles¦apply to these modern day processing plants. We show that Little's Law on long-term performance averages holds¦to computing centres, i.e. work-in-progress equals throughput rate multiplied by process lead time. Contrary to traditional¦manufacturing principles, the law of variation does not hold to computing centres, as the more variation in job lead¦times the better the throughput and utilisation of the system. We also show that as the utilisation of the system increases¦lead times and work-in-progress increase, which complies with traditional manufacturing. In comparison with current¦computing centre operations these results imply that better allocation of jobs could increase throughput and utilisation,¦while less computing resources are needed, thus increasing the overall efficiency of the centre. From a theoretical point¦of view, in a system with close to zero set-up times, as in the case of computing centres, the law of variation does not¦hold. We observe that the more variation in job lead times and resource usage, the higher the throughput of the system.
Vatter T. & Chavez-Demoulin V. (2015). Generalized Additive Models for Conditional Dependence Structures. Journal of Multivariate Analysis, 141, 147-167. [doi] [pdf] [web of science] [abstract]
We develop a generalized additive modeling framework for taking into account the effect of predictors on the dependence structure between two variables. We consider dependence or concordance measures that are solely functions of the copula, because they contain no marginal information: rank correlation coefficients or tail-dependence coefficients represent natural choices. We propose a maximum penalized log-likelihood estimator, derive its n-consistency and asymptotic normality, discuss details of the estimation procedure and the selection of the smoothing parameter. Finally, we present the results from a simulation study and apply the new methodology to a real dataset. Using intraday asset returns, we show that an intraday dependence pattern, due to the cyclical nature of market activity, is shaped similarly to the individual conditional second moments.
Vatter T., Wu H.-T., Chavez-Demoulin V. & Yu B. (2015). Non-parametric estimation of intraday spot volatility: disentangling instantaneous trend and seasonality. Econometrics, 3(4), 864-887. [doi] [abstract]
We provide a new framework for modeling trends and periodic patterns in high-frequency financial data. Seeking adaptivity to ever-changing market conditions, we enlarge the Fourier flexible form into a richer functional class: both our smooth trend and the seasonality are non-parametrically time-varying and evolve in real time. We provide the associated estimators and use simulations to show that they behave adequately in the presence of jumps and heteroskedastic and heavy-tailed noise. A study of exchange rate returns sampled from 2010 to 2013 suggests that failing to factor in the seasonality's dynamic properties may lead to misestimation of the intraday spot volatility.
Chavez-Demoulin V., Embrechts P. & Sardy S. (2014). Extreme-quantile tracking for financial time series. Journal of Econometrics, 181(1), 44-52. [doi] [pdf] [url] [web of science] [abstract]
Time series of financial asset values exhibit well-known statistical features such as heavy tails and volatility clustering. We propose a nonparametric extension of the classical Peaks-Over-Threshold method from extreme value theory to fit the time varying volatility in situations where the stationarity assumption may be violated by erratic changes of regime, say. As a result, we provide a method for estimating conditional risk measures applicable to both stationary and nonstationary series. A backtesting study for the UBS share price over the subprime crisis exemplifies our approach.
de Treville S., Bicer I., Chavez-Demoulin V., Hagspiel V., Schurhoff N., Tasserit C. & Wager S. (2014). Valuing lead time. Journal of Operations Management, 32(6), 337-346. [doi] [pdf] [web of science] [abstract]
When do short lead times warrant a cost premium? Decision makers generally agree that short lead times enhance competitiveness, but have struggled to quantify their benefits. Blackburn (2012) argued that the marginal value of time is low when demand is predictable and salvage values are high. de Treville et al. (2014) used real-options theory to quantify the relationship between mismatch cost and demand volatility, demonstrating that the marginal value of time increases with demand volatility, and with the volatility of demand volatility. We use the de Treville et al. model to explore the marginal value of time in three industrial supply chains facing relatively low demand volatility, extending the model to incorporate factors such as tender-loss risk, demand clustering in an order-up-to model, and use of a target fill rate that exceeded the newsvendor profit-maximizing order quantity. Each of these factors substantially increases the marginal value of time. In all of the companies under study, managers had underestimated the mismatch costs arising from lead time, so had underinvested in cutting lead times.
Appelqvist P., Chavez-Demoulin V., Hameri A.-P., Heikkiläc J. & Wautersa V. (2013). Turnaround across diverse global supply chains using shared metrics and change methodology: The Case of Amer Sports Corporation. International Journal of Operations and Production Management, 33(5), 622-647. [doi] [pdf] [url] [web of science] [abstract]
Purpose - The purpose of this paper is to document the outcome of a global three-year long supply chain improvement initiative at a multi-national producer of branded sporting goods that is transforming from a holding structure to an integrated company. The case company is comprised of seven internationally well-known sport brands, which form a diverse set of independent sub-cases, on which the same supply chain metrics and change project approach was applied to improve supply chain performance.¦Design/methodology/approach - By using in-depth case study and statistical analysis the paper analyzes across the brands how supply chain complexity (SKU count), supply chain type (make or buy) and seasonality affect completeness and punctuality of deliveries, and inventory as the change project progresses.¦Findings - Results show that reduction in supply chain complexity improves delivery performance, but has no impact on inventory. Supply chain type has no impact on service level, but brands with in-house production are better in improving inventory than those with outsourced production. Non-seasonal business units improve service faster than seasonal ones, yet there is no impact on inventory.¦Research limitations/implications - The longitudinal data used for the analysis is biased with the general business trend, yet the rich data from different cases and three-years of data collection enables generalizations to a certain level.¦Practical implications - The in-depth case study serves as an example for other companies on how to initiate a supply chain improvement project across business units with tangible results.¦Originality/value - The seven sub-cases with their different characteristics on which the same improvement initiative was applied sets a unique ground for longitudinal analysis to study supply chain complexity, type and seasonality.
Chavez-Demoulin V. & Davison A. C. (2012). Modelling time series extremes. REVSTAT - Statistical Journal, 10(1), 109-133. [url] [web of science] [abstract]
The need to model rare events of univariate time series has led to many recent advances in theory and methods. In this paper, we review telegraphically the literature on extremes of dependent time series and list some remaining challenges.
Chavez-Demoulin V. & McGill J. A. (2012). High-frequency financial data modeling using Hawkes processes. Journal of Banking and Finance, 36(12), 3415-3426. [doi] [pdf] [url] [abstract]
Intraday Value-at-Risk (VaR) is one of the risk measures used by market participants involved in high-frequency trading. High-frequency log-returns feature important kurtosis (fat tails) and volatility clustering (extreme log-returns appear in clusters) that VaR models should take into account. We propose a marked point process model for the excesses of the time series over a high threshold that combines Hawkes processes for the exceedances with a generalized Pareto distribution model for the marks (exceedance sizes). The conditional approach features intraday clustering of extremes and is used to calculate instantaneous conditional VaR. The models are backtested on real data and compared to a competitor approach that proposes a nonparametric extension of the classical peaks-over-threshold method. Maximum likelihood estimation is computationally intensive; we use a differential evolution genetic algorithm to find adequate starting values for the optimization process.
Chavez-Demoulin V., Davison A. C. & Frossard L. (2011). Discussion of the paper: Threshold modelling of spatially dependent non-stationary extremes with application to hurricane-induced wave heights. Environmetrics, 22(7), 810-816. [doi] [url] [web of science]
Chavez-Demoulin V. & Embrechts P. (2011). Operational risk. Encyclopedia of Quantitative Finance. [doi] [url] [abstract]
Under the capital requirements of the Basel II regime, banks have to provide estimates of their operational risk on a yearly basis and, for larger international banks, use a Value-at- Risk estimate at the confidence level of 99.9%. The so-called loss distribution approach (LDA) allows for broad methodological flexibility for the estimation of this risk capital. In this article, we discuss some of the underlying statistical issues.
Chavez-Demoulin V. & Embrechts P. (2010). Copulas in insurance. Encyclopedia of Quantitative Finance, 379-382. [doi] [url] [abstract]
The modeling of dependence has increasingly become one of the key issues in insurance. We highlight the use as well as the misuse of the concept of copula in this context. The main themes are the use of copula methodology in quantitative risk management and actuarial ruin theory.
Chavez-Demoulin V. & Embrechts P. (2010). Revisiting the edge, ten years on. Communications in Statistics - Theory and Methods, 39(8-9), 1674-1688. [doi] [url] [web of science] [abstract]
When these lines are written, it is January 21, 2008, a further "Black Monday" on the international markets. Stock indices have fallen between 5 and 10%. Which statistical tools help in describing such events and may help in understanding the consequences? In this article we update our knowledge on the modeling of extremal events, in particular with a view toward applications to finance, insurance, and risk management.
Chavez-Demoulin V., Embrechts P. & Neslehova J. (2006). Quantitative models for operational risk: extremes, dependence and aggregation. Journal of Banking and Finance, 30(10), 2635-2658. [doi] [pdf] [url] [web of science] [abstract]
Due to the new regulatory guidelines known as Basel II for banking and Solvency 2 for insurance, the financial industry is looking for qualitative approaches to and quantitative models for operational risk. Whereas a full quantitative approach may never be achieved, in this paper we present some techniques from probability and statistics which no doubt will prove useful in any quantitative modelling environment. The techniques discussed are advanced peaks over threshold modelling, the construction of dependent loss processes and the establishment of bounds for risk measures under partial information, and can be applied to other areas of quantitative risk management.
Chavez-Demoulin V., Embrechts P. & Neslehova J. (2006). Infinite mean models and the LDA for operational risk. Journal of Operational Risk, 1(1), 3-25. [web of science] [abstract]
Due to published statistical analyses of operational risk data, methodological approaches to the "advanced measurement approach" modeling of operational risk can be discussed in more detail. In this paper we raise some issues concerning correlation (or diversification) effects, the use of extreme value theory and the overall quantitative risk management consequences of extremely heavy-tailed data. We especially highlight issues around infinite-mean models. In addition to methodological examples and simulation studies, the paper contains indications for further research.
Chavez-Demoulin V. & Davison A. C. (2005). Generalized additive modelling of sample extremes. Journal of the Royal Statistical Society; Series C (Applied Statistics), 54(1), 207-222. [doi] [pdf] [url] [web of science] [abstract]
We describe smooth non-stationary generalized additive modelling for sample extremes, in which spline smoothers are incorporated into models for exceedances over high thresholds. Fitting is by maximum penalized likelihood estimation, with uncertainty assessed by using differences of deviances and bootstrap simulation. The approach is illustrated by using data on extreme winter temperatures in the Swiss Alps, analysis of which shows strong influence of the north Atlantic oscillation. Benefits of the new approach are flexible and appropriate modelling of extremes, more realistic assessment of estimation uncertainty and the accommodation of complex dependence patterns.
Chavez-Demoulin V., Davison A. C. & McNeil A. J. (2005). Estimating value-at-risk: a point process approach. Quantitative Finance, 5(2), 227-234. [doi] [pdf] [url] [web of science] [abstract]
We consider the modelling of extreme returns in financial time series, and introduce a marked point process model for the exceedances of a high threshold. This model has a self-exciting, Hawkes-process structure in which recent events affect the current intensity of threshold exceedances more than distant ones. Estimates of value-at-risk are derived for real datasets and the success of the estimation method is evaluated in backtests.
Chavez-Demoulin V. & Embrechts P. (2004). Smooth extremal models in finance and insurance. Journal of Risk and Insurance, 71(2), 183-199. [doi] [pdf] [url] [web of science] [abstract]
This article describes smooth nonstationary generalized additive modeling for sample extremes, in which spline smoothers are incorporated into models for exceedances over high thresholds. We summarize the smoothing methodology as a new tool for practical extreme value exploration in finance and insurance.
Chavez-Demoulin V., Embrechts P. & Roehrl A. (2002). A statistical analysis of the share price of the SAIR group (1996-2001) from a risk manager's point of view. Derivatives Use Trading and Regulation, 8(2), 105-122. [abstract]
Over recent years, extreme value theory (EVT). has been used in order to analyse statistically financial data showing clear non~normal behaviour Several examples in the areas of market, credit and operational risk have been discussed. This paper loo oks at the particular case of Swissair and quantifies, using EVT, the extremal behaviour of the returns. For this, the paper goes beyond traditional EVT and introduces new methodology such as smoothing and more advanced maximum likelihood techniques.
Chavez-Demoulin V. (1999). Bayesian inference for small-sample capture-recapture data. Biometrics, 55(3), 727-731. [doi] [pdf] [web of science] [abstract]
We consider data on the survival of a population of Cephalorhynchus hectori, Hector's dolphins, in a marine area of New Zealand. To estimate survival probabilities of animal populations, a multiple capture-recapture sampling scheme can be used. In this paper, we propose a practical methodology to derive approximations to posterior distributions based on Laplace methods. We show how to calculate Bayes estimates and credible intervals in this setting.
Compte-rendu Chavez-Demoulin V., Das B. & Embrechts P. (2010). Probabilistic Analysis of flooding at Murgenthal for Kernkraft-Goesgen Daeniken (KKG). RiskLab internal report.
Vulgarisation Chavez-Demoulin V. (2004). Was ist Extremwerttheorie ?. Risknews.
Chavez-Demoulin V. & Roehrl A. (2004). EVT can save your neck (?). Bulletin of Swiss Statistical Society.
Chavez-Demoulin V., Roehrl A. S. A., Roehrl R. A. & Schmiedl S. W. (2002). Datamining mit R. Linux-Enterprise, 2.
Chavez-Demoulin V., Weinberg A., Berezka V., Roehrl A. & Schmiedl S. W. (2002). Risk reduction : Transparent real-time enterprise. Banks and Technologies, 9.
Parties de livre
Chapitre Chavez-Demoulin V. & Embrechts P. (2011). An EVT primer for credit risk. In Lipton, A. & Rennie, A. (Ed.), The Oxford Handbook of Credit Derivatives (pp. 500-532). Oxford University Press. [doi] [abstract]
This article aims to provide the basics any risk manager should know on the modelling of external events, and this from a past-present-future research perspective. Such events are often also referred to as low-probability events or rare events. The article is organised as follows. Section 2 starts with an overview of the credit risk-specific issues within Quantitative Risk Management and shows where relevant Extreme Value Theory-related questions are being asked. Section 3 presents the one-dimensional theory of extremes, whereas Section 4 is concerned with the multivariate case. Section 5 discusses particular applications and gives an outlook on current research in the field, while Section 6 concludes.
Sardy S., Bilat C., Tseng P. & Chavez-Demoulin V. (2002). A comparison between L1 Markov random field-based and wavelet-based estimators. In Dodge Y. (Ed.), Statistical Data Analysis Based on the L1-Norm and Related Methods (pp. 395-404). Birkhäuser Verlag.
Actes de conférence (partie) Chavez-Demoulin V., Jarvis S., Perera R., Roehrl A., Schmiedl S. & M.P. Sondergaard (2003). Extreme Datamining. In Schader M., Gaul W. & Vichi M. (Eds.), Proceedings of the 26th Annual Conference of the Gesellschaft für Klassifikation e.V., University of Mannheim, July 2002, Between Data Science and Applied Data Analysis (pp. 387-394). Springer.
Rapports Chavez-Demoulin V. & Davison A. C. (2010). Statistics of hydrological extreme values : Exploratory analysis, modelling and recommendations. Bundesamt für Umwelt.
Thèses Chavez-Demoulin V., Davison A. C. (Dir.) (1999). Two Problems in Environmental Statistics : Capture-Recapture Analysis and Smooth Extremal models. EPFL.