Corpus ID: 152559102

Incorporating fat tails in financial models using entropic divergence measures

  title={Incorporating fat tails in financial models using entropic divergence measures},
  author={Santanu S. Dey and Sandeep Juneja},
  journal={arXiv: Statistical Finance},
In the existing financial literature, entropy based ideas have been proposed in portfolio optimization, in model calibration for options pricing as well as in ascertaining a pricing measure in incomplete markets. The abstracted problem corresponds to finding a probability measure that minimizes the relative entropy (also called $I$-divergence) with respect to a known measure while it satisfies certain moment constraints on functions of underlying assets. In this paper, we show that under $I… Expand

Figures and Tables from this paper

On Optimization over Tail Distributions
This work provides a technique in parallel to Choquet's theory, via a combination of integration by parts and change of measures, to transform shape constrained problems into families of moment problems, and shows how a moment problem cast over infinite support can be reformulated into a problem over compact support with an additional slack variable. Expand
Engineering More Effective Weighted Monte Carlo Option Pricing Models
The Weighted Monte Carlo (WMC) method (Avellaneda (1998) and Avellaneda et al. (2001)) is a practical and robust method for calibrating a contingent claim pricing model that is consistent with theExpand
The Diversification Delta: A Different Perspective
In a 2012 article published in The Journal of Portfolio Management, Vermorken, Medda, and Schröder introduce a new measure of diversification, the Diversification Delta (DD), based on the entropy ofExpand
Robust bounds in multivariate extremes
Extreme value theory provides an asymptotically justified framework for estimation of exceedance probabilities in regions where few or no observations are available. For multivariate tail estimation,Expand
Estimating Multivariate Conditional Models via Entropic Methods
We introduce a new practical numerical method to estimate conditional distributions, p(y|x), where y is the value of a continuous random variable supported on R^{N_y} and x is in R^{N_x}, via theExpand
Parametric Stress-Testing in Non-Normal Markets via Copula-Marginal Entropy Pooling
A novel approach for stress-testing (portfolios of) financial assets is presented. The technique extends the parametric Entropy Pooling approach to skewed and thick-tailed markets. The techniqueExpand
Robust Sensitivity Analysis for Stochastic Systems
  • H. Lam
  • Mathematics, Computer Science
  • Math. Oper. Res.
  • 2016
This work studies a worst-case approach to measure the sensitivity to model misspecification in the performance analysis of stochastic systems, and develops infinitesimal approximations for these programs, resulting in asymptotic expansions of their optimal values as the divergence shrinks to zero. Expand
Robust Actuarial Risk Analysis
Techniques for the assessment of model error in the context of insurance risk analysis are investigated, based on finding robust estimates for actuarial quantities of interest obtained by solving optimization problems over the unknown probabilistic models. Expand
Optimization-based Quantification of Simulation Input Uncertainty via Empirical Likelihood
We study an optimization-based approach to construct statistically accurate confidence intervals for simulation performance measures under nonparametric input uncertainty. This approach computesExpand
Advanced tutorial: Input uncertainty and robust analysis in stochastic simulation
  • H. Lam
  • Computer Science
  • 2016 Winter Simulation Conference (WSC)
  • 2016
This tutorial overviews some methodological developments in major established statistical methods and discusses some recent results from a robust-optimization-based viewpoint and their comparisons to the established methods. Expand


Minimum-Relative-Entropy Calibration of Asset-Pricing Models
We present an algorithm for calibrating asset-pricing models to the prices of benchmark securities. The algorithm computes the probability that minimizes the relative entropy with respect to a priorExpand
A general approach for calibrating Monte Carlo models to the market prices of benchmark securities is presented. Starting from a given model for market dynamics (price diffusion, rate diffusion,Expand
Conditional Distribution in Portfolio Theory
We present a new method to obtain a conditional mean vector and a conditional covariance matrix when given an investor's view about return profiles of certain assets. The method extends earlierExpand
Global Portfolio Optimization Revisited : A Least Discrimination Alternative to Black-Litterman ICMA Centre Discussion Papers in Finance DP 2007-07 This Version : 7 th July 2007
Global portfolio optimization models rank among the proudest achievements of modern finance theory, but practitioners are still struggling to put them to work. In 1992, Black and Litterman recognizedExpand
Asset Allocation
nvestors create global bond portfolios for a variety of reasons: to diversify interest rate risk, to manage yield, to control exposure to foreign currencies, and to enlarge the universe of possibleExpand
Abstract In this paper we study derivative pricing under information on observed market prices of some derivatives. To this purpose we characterize minimal distance martingale measures underExpand
Private Firm Default Probabilities Via Statistical Learning Theory and Utility Maximization
We estimate real-world private firm default probabilities over a fixed time horizon,conditioned on a vector of explanatory variables, which include financial ratios, economic indicators, and marketExpand
Return to RiskMetrics: The Evolution of a Standard
It has been almost five years since the formal statement of the model was updated, so a new document is created, Return to RiskMetrics: The Evolution of a Standard, which updates and restatement of the mathematical models in the 1996 Riskmetrics Technical Document. Expand
The Minimal Entropy Martingale Measure and the Valuation Problem in Incomplete Markets
Let χ be a family of stochastic processes on a given filtered probability space (Ω, "F", ("F" "t" ) "t" is an element of "T" , "P") with "T"⊆R p . Under the assumption that the set "M" "e" ofExpand
A Simple Nonparametric Approach to Derivative Security Valuation
Canonical valuation uses historical time series to predict the probability distribution of the discounted value of primary assets' discounted prices plus accumulated dividends at any future date.Expand