Jeudi 28 Novembre 2013 - journée ISFA-IRA
Programme journée ISFA-IRA
9h - 9h15 : Accueil et Café
9h15 - 10h : Nicole El Karoui (UPMC) : Fast Detection on Proportional Two-Population Hazard Rates
10h - 10h45 : Stéphane Loisel (ISFA) : ANR project Lolita : Longevity with Life style Improvements
10h45 - 11h15 : Pause café et discussion libre
11h15 - 12h : Fréderic Karamé (GAINS-IRA) : Hamilton Smooth Particle Filters
12h - 14h : Déjeuner
14h - 14h45 : Ying Jiao (ISFA) : Hedging under multiple risk constraints
14h45 - 15h30 : Alexandre Brouste (LMM-IRA) : Parametric statistical inference in fractional Ornstein-Uhlenbeck process
15h30 - 16h15 : Christian Robert (ISFA) : Distortion risk measures, ambiguity aversion and optimal effort
16h15 - 17h : Café et discussion libre
Alexandre BROUSTE (LMM-IRA) : Parametric statistical inference in fractional Ornstein-Uhlenbeck process
Abstract : Several statistical models that imply the fractional Ornstein-Uhlenbeck (fOU) process will be presented : direct observations of the process or partial observations in an additive independent noise, continuous observations or discrete observations. In this different settings, we exhibit large sample (or high-frequency) asymptotical properties of the estimators (maximum likelihood estimator, quadratic variation based estimator, moment estimator, …) for all parameters of interest of the fOU. We also illustrate our results with the R package yuima.
Nicole EL KAROUI (UPMC) : Fast Change Detection on Proportional Two-Population Hazard Rates
Abstract : We consider the problem of detecting an abrupt change on the structural relationship between two populations using a proportional hazard framework. We broadly follow the robust framework introduced in [?] adapted to the case of sequentially observed counting processes. We use the so-called cumulative sums (cusum) procedure in order to sound an alarm as quick as the change occurs. Such a problem is very similar to the ruin problem of an insurance company paying dividends proportional to the reserves.
In this paper, we propose a simplified proof of the optimality of the cusum rule based on some well-suited martingales but for a different performance criteria. Closed form formula for the performance of the procedure are also derived. Finally, we draw some numerical tests to assess the effectiveness of the procédure (joint work with S.Loisel, C. Mazza, Y.Sahli).
Ying JIAO (ISFA) : Hedging under multiple risk constraints
Abstract : Motivated by the asset-liability management of a nuclear power plant operator, we consider the problem of finding the least expensive portfolio, which outperforms a given set of stochastic benchmarks. We consider different alternative formulations of this problem in a complete market setting, establish the relationship between these formulations, present a general resolution methodology via dynamic programming in a non-Markovian context and give explicit solutions in special cases.
Fréderic KARAME (GAINS-IRA) : Hamilton Smooth Particle Filters
Abstract : We propose a new particle filtering approach to estimate the Markov-switching stochastic volatility model with leverage. Its structure is based on the Hamilton filter. We use a sequential importance sampling particle filter to approximate the unobserved log-volatility and calculate the conditional likelihood necessary for the regime probabilities update and parameters estimation. In order to easily update particles and implement likelihood-based inference, we use the smooth resampling approach developed by Malik & Pitt (2011a). Our last contribution relies on the choice of the proposal distribution for the importance sampling step of the particle filter. After the description of our approach and some simulation experiments, we present the estimation results for the IBOVESPA index for comparison with Carvalho & Lopes (2007).
Stéphane LOISEL (ISFA) : ANR project Lolita : Longevity with Life style Improvements
Abstract : We present the ANR project Lolita and early research related to this project.
Christian Robert (ISFA) : Distortion risk measures, ambiguity aversion and optimal effort
Abstract : We consider the class of concave distortion risk measures to study how choice is influenced by the decision-maker’s attitude to risk and provide comparative statics results. We also assume ambiguity about the probability distribution of the risk and consider a framework à la Klibanoff, Marinacci and Mukerji (2005) to study the value of information that resolves ambiguity. We show that this value increases with greater ambiguity, with greater ambiguity aversion, and in some cases with greater risk aversion. Finally we examine whether a more risk-averse and a more ambiguity-averse individual will invest in more effort to shift his initial risk distribution to a better target distribution.