Search results for "Gaussian distribution"
showing 10 items of 16 documents
Portfolio performance evaluation with generalized Sharpe ratios: Beyond the mean and variance
2009
The main purpose of this paper is to present a theoretically sound portfolio performance measure that takes into account higher moments of the distribution of returns. First, we perform a study of the investor's preferences to higher moments of distribution within expected utility theory and discuss the performance measurement. To illustrate the investor's preferences to higher moments and the computation of a performance measure, we provide an approximation analysis of the optimal capital allocation problem and derive a formula for the Sharpe ratio adjusted for skewness of distribution. This performance measure justifies the notion of the Generalized Sharpe Ratio (GSR) introduced by Hodges…
Modeling Term Structure Dynamics in the Nordic Electricity Swap Market
2010
We analyze the daily returns of Nordic electricity swaps and identify significant risk premia in the short end of the market. On average, long positions in this part of the swap market yield negative returns. The daily returns are distinctively non-normal in terms of tail-fatness, but we find little evidence of asymmetry. We investigate if the flexible four-parameter class of normal inverse Gaussian (NIG) distributions can capture the observed stylized facts and find that this class of distributions offers a remarkably improved fit relative to the normal distribution. We also compare the fit with that of the four-parameter class of stable distributions; the NIG law outperforms the stable la…
A New Nonparametric Estimate of the Risk-Neutral Density with Applications to Variance Swaps
2021
We develop a new nonparametric approach for estimating the risk-neutral density of asset prices and reformulate its estimation into a double-constrained optimization problem. We evaluate our approach using the S\&P 500 market option prices from 1996 to 2015. A comprehensive cross-validation study shows that our approach outperforms the existing nonparametric quartic B-spline and cubic spline methods, as well as the parametric method based on the Normal Inverse Gaussian distribution. As an application, we use the proposed density estimator to price long-term variance swaps, and the model-implied prices match reasonably well with those of the variance future downloaded from the CBOE websi…
Separate regression modelling of the Gaussian and Exponential components of an EMG response from respiratory physiology.
2014
If Y1 \sim N(\mu ;\sigma^2) and Y2 \sim Exp(\nu), with Y1 independent of Y2, then their sum Y = Y1 +Y2 follows an Exponentially Modified Gaussian (EMG) distribution. In many applications it is of interest to model the two components separately, in order to investigate their (possibly) different important predictors. We show how this can be done through a GAMLSS with EMG response, and apply this separate regression modelling strategy to a dataset on lung function variables from the SAPALDIA cohort study.
Stochastic dynamical modelling of spot freight rates
2014
Based on empirical analysis of the Capesize and Panamax indices, we propose different continuous-time stochastic processes to model their dynamics. The models go beyond the standard geometric Brownian motion, and incorporate observed effects like heavy-tailed returns, stochastic volatility and memory. In particular, we suggest stochastic dynamics based on exponential Levy processes with normal inverse Gaussian distributed logarithmic returns. The Barndorff-Nielsen and Shephard stochastic volatility model is shown to capture time-varying volatility in the data. Finally, continuous-time autoregressive processes provide a class of models sufficiently rich to incorporate short-term persistence …
Observation and Mass Measurement of the BaryonΞb−
2007
We report the observation and measurement of the mass of the bottom, strange baryon $\Xi^-_b$ through the decay chain $\Xi^-_b \to J/\psi \Xi^-$, where $J/\psi \to \mu^+ \mu^-$, $\Xi^- \to \Lambda \pi^-$, and $\Lambda \to p \pi^-$. Evidence for observation is based on a signal whose probability of arising from the estimated background is 6.6 x 10^{-15}, or 7.7 Gaussian standard deviations. The $\Xi^-_b$ mass is measured to be $5792.9\pm 2.5$ (stat.) $\pm 1.7$ (syst.) MeV/$c^2$.
Tool-life modelling as a stochastic process
1998
Abstract In a previous paper [G. Galante, A. Lombardo, A. Passannanti, Proceedings of XXXVII Scientific Meeting of the Italian Statistical Society, 1994, p. 553] the Authors proposed to model cutting tool wear behaviour as a stochastic process with independent Gaussian increments plus drift. Such a model implies that the tool-life, i.e. the time to reach a fixed value of flank wear, has an inverse Gaussian probability distribution. The model has several practical and theoretical advantages. In fact, it is based on an easily and cheaply experimentally verifiable wear behaviour hypothesis, it is more flexible because it is not limited to a particular wear level and, finally, the data are bett…
On VaR using modified gaussian copula
2008
The problem of modeling asset returns is one of the most important issue in finance. People generally use Gaussian processes because of their tractable properties for computation. However, it is well known that asset returns are fat-tailed leading to an underestimation of the risk. One of the most recent proposals is to model the interdependence of asset returns, for example in a portfolio, by means of Copulas and choose marginal distributions with fat tail to fit the single asset returns. The aim of the paper is to show first results concerning the evaluation of Portfolio Value-at-Risk (VaR) using the Gaussian copula, modified by introducing a particular correlation coefficient, and assumi…
Discrete Time Portfolio Selection with Lévy Processes
2007
This paper analyzes discrete time portfolio selection models with Lévy processes. We first implement portfolio models under the hypotheses the vector of log-returns follow or a multivariate Variance Gamma model or a Multivariate Normal Inverse Gaussian model or a Brownian Motion. In particular, we propose an ex-ante and an ex-post empirical comparisons by the point of view of different investors. Thus, we compare portfolio strategies considering different term structure scenarios and different distributional assumptions when unlimited short sales are allowed.
A Comparison among Portfolio Selection Strategies with Subordinated Lévy Processes
2007
In this paper we describe portfolio selection models using Lévy processes. The contribution consists in comparing some portfolio selection strategies under different distributional assumptions. We first implement portfolio models under the hypothesis the log-returns follow a particular process with independent and stationary increments. Then we compare the ex-post final wealth of optimal portfolio selection models with subordinated Lévy processes when limited short sales and transaction costs are allowed.