2010, Vol. 34, Núm. 1
http://hdl.handle.net/2099/10746
Sun, 20 Oct 2019 12:36:35 GMT2019-10-20T12:36:35ZVariance reduction technique for calculating value at risk in fixed income portfolios
http://hdl.handle.net/2099/11222
Variance reduction technique for calculating value at risk in fixed income portfolios
Abad, Pilar; Benito, Sonia
Financial institutions and regulators increasingly use Value at Risk (VaR) as a standard measure
for market risk. Thus, a growing amount of innovative VaR methodologies is being developed by
researchers in order to improve the performance of traditional techniques. A variance-covariance
approach for fixed income portfolios requires an estimate of the variance-covariance matrix of
the interest rates that determine its value. We propose an innovative methodology to simplify the
calculation of this matrix. Specifically, we assume the underlying interest rates parameterization
found in the model proposed by Nelson and Siegel (1987) to estimate the yield curve. As this paper
shows, our VaR calculating methodology provides a more accurate measure of risk compared to
other parametric methods
Fri, 28 Oct 2011 13:12:32 GMThttp://hdl.handle.net/2099/112222011-10-28T13:12:32ZAbad, PilarBenito, SoniaFinancial institutions and regulators increasingly use Value at Risk (VaR) as a standard measure
for market risk. Thus, a growing amount of innovative VaR methodologies is being developed by
researchers in order to improve the performance of traditional techniques. A variance-covariance
approach for fixed income portfolios requires an estimate of the variance-covariance matrix of
the interest rates that determine its value. We propose an innovative methodology to simplify the
calculation of this matrix. Specifically, we assume the underlying interest rates parameterization
found in the model proposed by Nelson and Siegel (1987) to estimate the yield curve. As this paper
shows, our VaR calculating methodology provides a more accurate measure of risk compared to
other parametric methodsA family of ratio estimators for population mean in extreme ranked set sampling using two auxiliary variables
http://hdl.handle.net/2099/11221
A family of ratio estimators for population mean in extreme ranked set sampling using two auxiliary variables
Haq, Abdul; Shabbir, Javid
In this paper we have adopted the Khoshnevisan et al. (2007) family of estimators to extreme
ranked set sampling (ERSS) using information on single and two auxiliary variables. Expressions
for mean square error (MSE) of proposed estimators are derived to first order of approximation.
Monte Carlo simulations and real data sets have been used to illustrate the method. The results
indicate that the estimators under ERSS are more efficient as compared to estimators based on
simple random sampling (SRS), when the underlying populations are symmetric.
Fri, 28 Oct 2011 13:11:27 GMThttp://hdl.handle.net/2099/112212011-10-28T13:11:27ZHaq, AbdulShabbir, JavidIn this paper we have adopted the Khoshnevisan et al. (2007) family of estimators to extreme
ranked set sampling (ERSS) using information on single and two auxiliary variables. Expressions
for mean square error (MSE) of proposed estimators are derived to first order of approximation.
Monte Carlo simulations and real data sets have been used to illustrate the method. The results
indicate that the estimators under ERSS are more efficient as compared to estimators based on
simple random sampling (SRS), when the underlying populations are symmetric.Small-sample inference about variance and its transformations
http://hdl.handle.net/2099/11220
Small-sample inference about variance and its transformations
Longford, N.T.
We discuss minimum mean squared error and Bayesian estimation of the variance and its
common transformations in the setting of normality and homoscedasticity with small samples, for
which asymptotics do not apply. We show that permitting some bias can be rewarded by greatly
reduced mean squared error. We apply borderline and equilibrium priors. The purpose of these
priors is to reduce the onus on the expert or client to specify a single prior distribution that would
capture the information available prior to data inspection. Instead, the (parametric) class of all
priors considered is partitioned to subsets that result in the preference for different actions. With
the family of conjugate inverse gamma priors, this Bayesian approach can be formulated in the
frequentist paradigm, describing the prior as being equivalent to additional observations.
Fri, 28 Oct 2011 13:09:32 GMThttp://hdl.handle.net/2099/112202011-10-28T13:09:32ZLongford, N.T.We discuss minimum mean squared error and Bayesian estimation of the variance and its
common transformations in the setting of normality and homoscedasticity with small samples, for
which asymptotics do not apply. We show that permitting some bias can be rewarded by greatly
reduced mean squared error. We apply borderline and equilibrium priors. The purpose of these
priors is to reduce the onus on the expert or client to specify a single prior distribution that would
capture the information available prior to data inspection. Instead, the (parametric) class of all
priors considered is partitioned to subsets that result in the preference for different actions. With
the family of conjugate inverse gamma priors, this Bayesian approach can be formulated in the
frequentist paradigm, describing the prior as being equivalent to additional observations.Modelling spatial patterns of distribution and abundance of mussel seed using structured additive regression models
http://hdl.handle.net/2099/11035
Modelling spatial patterns of distribution and abundance of mussel seed using structured additive regression models
Pata, María P.; Rodríguez-Álvarez, María Xosé; Lustres-Pérez, Vicente; Fernández Pulpeiro, Eugenio; Cadarso-Suárez, Carmen
As mussel farming depends on sources of natural mussel seed, knowledge of factors is required to regulate both the spatial distribution and abundance of this resource. These spatial patterns were modelled using Bayesian STructured Additive Regression (STAR) models for categorical
data, based on a mixed-model representation. We used Bayesian penalized splines for modelling the continuous covariate effects and a Markov random field prior for estimating the spatial effects.
Thu, 20 Oct 2011 19:00:16 GMThttp://hdl.handle.net/2099/110352011-10-20T19:00:16ZPata, María P.Rodríguez-Álvarez, María XoséLustres-Pérez, VicenteFernández Pulpeiro, EugenioCadarso-Suárez, CarmenAs mussel farming depends on sources of natural mussel seed, knowledge of factors is required to regulate both the spatial distribution and abundance of this resource. These spatial patterns were modelled using Bayesian STructured Additive Regression (STAR) models for categorical
data, based on a mixed-model representation. We used Bayesian penalized splines for modelling the continuous covariate effects and a Markov random field prior for estimating the spatial effects.New aging properties of the Clayton-Oakes model based on multivariate dispersion
http://hdl.handle.net/2099/10761
New aging properties of the Clayton-Oakes model based on multivariate dispersion
Arias-Nicolás, José Pablo; Mulero, Julio; Núñez-Barrera, Olga; Suárez-Llorens, Alfonso
In this work we present a recent definition of Multivariate Increasing Failure Rate (MIFR) based on the concept of multivariate dispersion. This new definition is an extension of the univariate characterization of IFR distributions under dispersive ordering of the residual lifetimes. We apply
this definition to the Clayton-Oakes model. In particular, we provide several conditions to order in the multivariate dispersion sense the residual lifetimes of random vectors with a dependence structure given by the Clayton-Oakes survival copula. We illustrate our results with a graphical
method.
Tue, 04 Oct 2011 18:13:03 GMThttp://hdl.handle.net/2099/107612011-10-04T18:13:03ZArias-Nicolás, José PabloMulero, JulioNúñez-Barrera, OlgaSuárez-Llorens, AlfonsoIn this work we present a recent definition of Multivariate Increasing Failure Rate (MIFR) based on the concept of multivariate dispersion. This new definition is an extension of the univariate characterization of IFR distributions under dispersive ordering of the residual lifetimes. We apply
this definition to the Clayton-Oakes model. In particular, we provide several conditions to order in the multivariate dispersion sense the residual lifetimes of random vectors with a dependence structure given by the Clayton-Oakes survival copula. We illustrate our results with a graphical
method.