Emmanuelle Jay

La Banque Postale Asset Management · DDDI

This paper presents how the use of a cleaned and robust covariance matrix estimate can improve significantly the overall performance of Maximum Variety and Minimum Variance portfolios. We assume that the asset returns are modelled through a multi-factor model where the error term is a multivariate and correlated elliptical symmetric noise extending...

This paper presents how the most recent improvements made on covariance matrix estimation and model order selection can be applied to the portfolio optimization problem. Our study is based on the case of the Maximum Variety Portfolio and may be obviously extended to other classical frameworks with analogous results. We focus on the fact that the as...

This paper presents how the most recent improvements made on covariance matrix estimation and model order selection can be applied to the portfolio optimization problem. Our study is based on the case of the Maximum Variety Portfolio and may be obviously extended to other classical frameworks with analogous results. We focus on the fact that the as...

Granger non-causality tests are widely used in empirical econometric analysis. Such tests are designed to assess the existence and direction of causal relationships, not the magnitude of causal relationships, although these are required in many applications. In the frequency domain, several approaches, such as the Direct Transfer Function measure,...

This paper presents how the most recent improvements made on covariance matrix estimation and model order selection can be applied to the portfolio optimisation problem. The particular case of the Maximum Variety Portfolio is treated but the same improvements apply also in the other optimisation problems such as the Minimum Variance Portfolio. We a...

This paper presents how the most recent improvements made on covariance matrix estimation and model order selection can be applied to the portfolio optimisation problem. The particular case of the Maximum Variety Portfolio is treated but the same improvements apply also in the other optimisation problems such as the Minimum Variance Portfolio. We a...

The twelve papers in this special issue presents relevant research contributions from the disciplines of finance, mathematics, data science and engineering to facilitate scientific cross-fertilization. It will also serve the signal processing community to be exposed to the state of the art in mathematical finance, financial engineering, financial s...

The standard mean-variance approach can imply extreme weights in some assets in the optimal allocation and a lack of stability of this allocation over time. In order to not only improve the robustness of the portfolio allocation, but also to better control the portfolio turnover and the sensitivity of the portfolio to systematic risk, it is propose...

This chapter introduces the common version of linear factor models and also discusses its limits and developments. It introduces different notations and discusses the model and its structure. The chapter lists out the reasons why factor models are generally used in finance, and further explains the limits of this approach. It also deals with the di...

This chapter focuses on the empirical ad hoc approach and presents three reference models that are widely used in the literature. These models are all based on the factor representation, but highlight the nature of the factors to be used to explain specific asset class returns. In a section, the authors denote by eigenfactors the factors obtained f...

This chapter introduces, illustrates and derives both least squares estimation (LSE) and Kalman filter (KF) estimation of the alpha and betas of a return, for a given number of factors that have already been selected. It formalizes the “per return factor model” and the concept of recursive estimate of the alpha and betas. The chapter explains the s...

This chapter presents a new family of algorithms named regularized Kalman Filters (rgKFs) that have been derived to detect and estimate exogenous outliers that might occur in the observation equation of a standard Kalman filter (KF). Inspired from the robust Kalman filter (RKF) of Mattingley and Boyd, which makes use of a l1-regularization step, th...

With recent outbreaks of multiple large-scale financial crises, amplified by interconnected risk sources, a new paradigm of fund management has emerged. This new paradigm leverages "embedded" quantitative processes and methods to provide more transparent, adaptive, reliable and easily implemented "risk assessment-based" practices. This book survey...

This paper presents a simple and efficient exogenous outlier detection & estimation algorithm introduced in a regularized version of the Kalman Filter (KF). Exogenous outliers that may occur in the observations are considered as an additional stochastic impulse process in the KF observation equation that requires a regularization of the innovation...

This chapter introduces the common version of linear factor models and also discusses its limits and developments. It introduces different notations and discusses the model and its structure. The chapter lists out the reasons why factor models are generally used in finance, and further explains the limits of this approach. It also deals with the di...

This chapter introduces, illustrates and derives both least squares estimation (LSE) and Kalman filter (KF) estimation of the alpha and betas of a return, for a given number of factors that have already been selected. It formalizes the “per return factor model” and the concept of recursive estimate of the alpha and betas. The chapter explains the s...

This chapter presents a new family of algorithms named regularized Kalman Filters (rgKFs) that have been derived to detect and estimate exogenous outliers that might occur in the observation equation of a standard Kalman filter (KF). Inspired from the robust Kalman filter (RKF) of Mattingley and Boyd, which makes use of a l1-regularization step, th...

This chapter focuses on the empirical ad hoc approach and presents three reference models that are widely used in the literature. These models are all based on the factor representation, but highlight the nature of the factors to be used to explain specific asset class returns. In a section, the authors denote by eigenfactors the factors obtained f...

The standard mean-variance approach can imply extreme weights in some assets in the optimal allocation and a lack of stability of this allocation over time. To improve the robustness of the portfolio allocation, but also to better control for the portfolio turnover and the sensitivity of the portfolio to systematic risk, it is proposed in this pape...

This paper presents a simple and efficient exogenous outlier detection & estimation algorithm introduced in a regularized version of the Kalman Filter (KF). Exogenous outliers that may occur in the observations are considered as an additional stochastic impulse process in the KF observation equation that requires a regularization of the innovation...

This article surveys the existing literature on the most widely used factor models employed in the realm of a financial asset pricing field. Through the concrete application of evaluating risks in the hedge fund industry, this article demonstrates that signal processing techniques are an interesting alternative to the selection of factors and can p...

We derive the expression of an optimum non-Gaussian radar detector from the non-Gaussian spherically invariant random process (SIRP) clutter model and a bayesian estimator of the SIRP characteristic density. SIRP modelizes non-Gaussian process as a complex Gaussian process whose variance, the so-called texture, is itself a positive random variable...

We propose to apply an efficient approximat ion saddle poin t approximation N to some fusion problems. The aim of this method is to approximate the law of a sum of (n) iid (independent identically distributed) variables. A lot of statistical problems (maximum likelihood estimator, hypothesis testing, ..) can be formulated as such a sum. This approx...

In this paper, a theoretical expression of the optimum non-Gaussian radar detector is derived from the non-Gaussian SIRP model (Spherically Invariant Random Process) clutter and a bayesian estimator of the characteristic function of the SIRP. The SIRP model is used to perform coherent detection and to modelize the clutter as a complex Gaussian proc...

For a long time, radar echoes coming from the various returns of the transmitted signal on many objects of the environment (clutter) have been exclusively modelled by Gaussian vectors. The related optimal detection procedure was then performed by the classical matched filter. Then, the technological improvement of radar systems showed that the true...

An expression for the optimum non-Gaussian radar detector is derived from the non-Gaussian SIRP model (spherically invariant random process) clutter and a Pade approximation of the characteristic function of the SIRP. The SIRP model is used to perform coherent detection and to modelize the non-Gaussian clutter as a complex Gaussian process whose va...

Original methods for radar detection performance analysis are derived for a fluctuating or non-fluctuating target embedded in additive and a priori unknown noise. This kind of noise can be, for example, the sea or ground clutter encountered in surface-based radar for the detection of low grazing angle targets and/or in high-resolution radar. In the...

The clutter encountered in low grazing an-gle situations is generally a non gaussian impulsive noise resulting in a mismatching of the classical radar detec-tor adjusted to the detection threshold of the gaussian hypothesis. To estimate the true detection performances of the radar, one has to take into account not only the power of the noise (as ma...

Original methods of radar detection performance analysis are derived for a fluctuating or non-fluctuating target embedded in additive and a priori unknown noise. This kind of noise can be, for example, the sea or ground clutter encountered in surface-sited radar for the detection of a target illuminated at low grazing angles or in high resolution r...

Dans le cadre de la détection d'une cible évoluant à site bas ou d'une cible éclairée par un radar à haute résolution distance, la nature impulsionnelle du bruit environnant écarte l'hypothèse gaussienne généralement retenue pour la modélisation de la statistique de ce fouillis. Nous proposons, dans ce papier, une modélisation par approximants de P...

This paper surveys the existing literature on the most widely-used factor models employed in the realm of financial asset pricing field. Through the concrete application of evaluating risks in the hedge fund industry, this paper demonstrates that signal processing techniques are an interesting alternative to the selection of factors and can provide...

In this paper detection performances of the Bayesian Optimum Radar Detector (BORD) against non-Gaussian ground clutter data are showed in the case of an unknown complex amplitude target. Recalling first how BORD was derived from the non-Gaussian SIRP model (Spherically Invariant Random Process) clutter we derive theoretical performances of the asym...

Les échos radar provenant des diverses réflexions du signal émis sur les éléments de l'environnement (le fouillis) ont longtemps été modélisés par des vecteurs Gaussiens. La procédure optimale de détection se résumait alors en la mise en oeuvre du filtre adapté classique. Avec l'évolution technologique des systèmes radar, la nature réelle du fouill...