Evaluating multivariate volatility forecasts : how effective are statistical and economic loss functions?

Multivariate volatility forecasts are an important input in many financial applications, in particular portfolio optimisation problems. Given the number of models available and the range of loss functions to discriminate between them, it is obvious that selecting the optimal forecasting model is challenging. The aim of this thesis is to thoroughly investigate how effective many commonly used statistical (MSE and QLIKE) and economic (portfolio variance and portfolio utility) loss functions are at discriminating between competing multivariate volatility forecasts. An analytical investigation of the loss functions is performed to determine whether they identify the correct forecast as the best forecast. This is followed by an extensive simulation study examines the ability of the loss functions to consistently rank forecasts, and their statistical power within tests of predictive ability. For the tests of predictive ability, the model confidence set (MCS) approach of Hansen, Lunde and Nason (2003, 2011) is employed. As well, an empirical study investigates whether simulation findings hold in a realistic setting. In light of these earlier studies, a major empirical study seeks to identify the set of superior multivariate volatility forecasting models from 43 models that use either daily squared returns or realised volatility to generate forecasts. This study also assesses how the choice of volatility proxy affects the ability of the statistical loss functions to discriminate between forecasts. Analysis of the loss functions shows that QLIKE, MSE and portfolio variance can discriminate between multivariate volatility forecasts, while portfolio utility cannot. An examination of the effective loss functions shows that they all can identify the correct forecast at a point in time, however, their ability to discriminate between competing forecasts does vary. That is, QLIKE is identified as the most effective loss function, followed by portfolio variance which is then followed by MSE. The major empirical analysis reports that the optimal set of multivariate volatility forecasting models includes forecasts generated from daily squared returns and realised volatility. Furthermore, it finds that the volatility proxy affects the statistical loss functions’ ability to discriminate between forecasts in tests of predictive ability. These findings deepen our understanding of how to choose between competing multivariate volatility forecasts.

US inflation expectations and heterogeneous loss functions, 1968-2010

Recent literature has suggested that macroeconomic forecasters may have asymmetric loss functions, and that there may be heterogeneity across forecasters in the degree to which they weigh under- and over-predictions. Using an individual-level analysis that exploits the Survey of Professional Forecasters respondents’ histogram forecasts, we find little evidence of asymmetric loss for the inflation forecasters

Fonctions de perte en actuariat

Mémoire numérisé par la Division de la gestion de documents et des archives de l'Université de Montréal

A Huber-loss-driven clustering technique and its application to robust cell detection in confocal microscopy images

We address the problem of detecting cells in biological images. The problem is important in many automated image analysis applications. We identify the problem as one of clustering and formulate it within the framework of robust estimation using loss functions. We show how suitable loss functions may be chosen based on a priori knowledge of the noise distribution. Specifically, in the context of biological images, since the measurement noise is not Gaussian, quadratic loss functions yield suboptimal results. We show that by incorporating the Huber loss function, cells can be detected robustly and accurately. To initialize the algorithm, we also propose a seed selection approach. Simulation results show that Huber loss exhibits better performance compared with some standard loss functions. We also provide experimental results on confocal images of yeast cells. The proposed technique exhibits good detection performance even when the signal-to-noise ratio is low.

Evaluating and classifying contaminated areas based on loss functions using annealing simulations

This paper presents a methodology based on geostatistical theory for quantifying the risks associated with heavy-metal contamination in the harbor area of Santana, Amapa State, Northern Brazil. In this area there were activities related to the commercialization of manganese ore from Serra do Navio. Manganese and arsenic concentrations at unsampled sites were estimated by postprocessing results from stochastic annealing simulations; the simulations were used to test different criteria for optimization, including average, median, and quantiles. For classifying areas as contaminated or uncontaminated, estimated quantiles based on functions of asymmetric loss showed better results than did estimates based on symmetric loss, such as the average or the median. The use of specific loss functions in the decision-making process can reduce the costs of remediation and health maintenance. The highest global health costs were observed for manganese. (c) 2008 Elsevier B.V. All rights reserved

Efficient algorithms for learning kernels from multiple similarity matrices with general convex loss functions

In this paper we consider the problem of learning an n × n kernel matrix from m(1) similarity matrices under general convex loss. Past research have extensively studied the m = 1 case and have derived several algorithms which require sophisticated techniques like ACCP, SOCP, etc. The existing algorithms do not apply if one uses arbitrary losses and often can not handle m > 1 case. We present several provably convergent iterative algorithms, where each iteration requires either an SVM or a Multiple Kernel Learning (MKL) solver for m > 1 case. One of the major contributions of the paper is to extend the well knownMirror Descent(MD) framework to handle Cartesian product of psd matrices. This novel extension leads to an algorithm, called EMKL, which solves the problem in O(m2 log n 2) iterations; in each iteration one solves an MKL involving m kernels and m eigen-decomposition of n × n matrices. By suitably defining a restriction on the objective function, a faster version of EMKL is proposed, called REKL,which avoids the eigen-decomposition. An alternative to both EMKL and REKL is also suggested which requires only an SVMsolver. Experimental results on real world protein data set involving several similarity matrices illustrate the efficacy of the proposed algorithms.

Quadratic Lyapunov functions for linear systems Journal

The paper deals with the existence of a quadratic Lyapunov function V = x′P(t)x for an exponentially stable linear system with varying coefficients described by the vector differential equation S0305004100044777_inline1 The derivative dV/dt is allowed to be strictly semi-(F) and the locus dV/dt = 0 does not contain any arc of the system trajectory. It is then shown that the coefficient matrix A(t) of the exponentially stable sy

on quadratic bent functions in polynomial forms

In this correspondence, we construct some new quadratic bent functions in polynomial forms by using the theory of quadratic forms over finite fields. The results improve some previous work. Moreover, we solve a problem left by Yu and Gong in 2006.

Designer aminoglycosides prevent cochlear hair cell loss and hearing loss.

Bacterial infections represent a rapidly growing challenge to human health. Aminoglycosides are widely used broad-spectrum antibiotics, but they inflict permanent hearing loss in up to ~50% of patients by causing selective sensory hair cell loss. Here, we hypothesized that reducing aminoglycoside entry into hair cells via mechanotransducer channels would reduce ototoxicity, and therefore we synthesized 9 aminoglycosides with modifications based on biophysical properties of the hair cell mechanotransducer channel and interactions between aminoglycosides and the bacterial ribosome. Compared with the parent aminoglycoside sisomicin, all 9 derivatives displayed no or reduced ototoxicity, with the lead compound N1MS 17 times less ototoxic and with reduced penetration of hair cell mechanotransducer channels in rat cochlear cultures. Both N1MS and sisomicin suppressed growth of E. coli and K. pneumoniae, with N1MS exhibiting superior activity against extended spectrum β lactamase producers, despite diminished activity against P. aeruginosa and S. aureus. Moreover, systemic sisomicin treatment of mice resulted in 75% to 85% hair cell loss and profound hearing loss, whereas N1MS treatment preserved both hair cells and hearing. Finally, in mice with E. coli-infected bladders, systemic N1MS treatment eliminated bacteria from urinary tract tissues and serially collected urine samples, without compromising auditory and kidney functions. Together, our findings establish N1MS as a nonototoxic aminoglycoside and support targeted modification as a promising approach to generating nonototoxic antibiotics.