Estimating value at risk with semiparametric support vector quantile regression

Value at Risk (VaR) has been used as an important tool to measure the market risk under normal market. Usually the VaR of log returns is calculated by assuming a normal distribution. However, log returns are frequently found not normally distributed. This paper proposes the estimation approach of VaR using semiparametric support vector quantile regression (SSVQR) models which are functions of the one-step-ahead volatility forecast and the length of the holding period, and can be used regardless of the distribution. We find that the proposed models perform better overall than the variance-covariance and linear quantile regression approaches for return data on S&P 500, NIKEI 225 and KOSPI 200 indices.     

Estimating the spectral measure of an extreme value distribution

Let (X₁, Y₁), (X₂, Y₂),…, (X_n, Y_n) be a random sample from a bivariate distribution function F which is in the domain of attraction of a bivariate extreme value distribution function G. This G is characterized by the extreme value indices and its spectral measure or angular measure. The extreme value indices determine both the marginals and the spectral measure determines the dependence structure. In this paper, we construct an empirical measure, based on the sample, which is a consistent estimator of the spectral measure. We also show for positive extreme value indices the asymptotic normality of the estimator under a suitable 2nd order strengthening of the bivariate domain of attraction condition.     

Measuring bank contagion in Europe using binary spatial regression models

The recent European sovereign debt crisis clearly illustrates the importance of measuring the contagion effects of bank failures. Indeed, to better understand and monitor contagion risk, the European Central Bank has assumed the supervision of the largest banks in each of the member states. We propose a measure of contagion risk based on the spatial autocorrelation parameter of a binary spatial autoregressive model. Using different specifications of the interbank connectivity matrix, we estimate the contagion parameter for banks within the Eurozone, between 1996 and 2012. We provide evidence of high levels of systemic risk due to contagion during the European sovereign debt crisis.     

Heavy-traffic extreme value limits for Erlang delay models

We consider the maximum queue length and the maximum number of idle servers in the classical Erlang delay model and the generalization allowing customer abandonment—the M/M/n+M queue. We use strong approximations to show, under regularity conditions, that properly scaled versions of the maximum queue length and maximum number of idle servers over subintervals [0,t] in the delay models converge jointly to independent random variables with the Gumbel extreme value distribution in the quality-and-efficiency-driven (QED) and ED many-server heavy-traffic limiting regimes as n and t increase to infinity together appropriately; we require that t _n →∞ and t _n =o(n ^1/2?ε ) as n→∞ for some ε>0.     

Conditional extreme value models: fallacies and pitfalls

Conditional extreme value models have been introduced by Heffernan and Resnick (Ann. Appl. Probab., 17, 537–571, 2007) to describe the asymptotic behavior of a random vector as one specific component becomes extreme. Obviously, this class of models is related to classical multivariate extreme value theory which describes the behavior of a random vector as its norm (and therefore at least one of its components) becomes extreme. However, it turns out that this relationship is rather subtle and sometimes contrary to intuition. We clarify the differences between the two approaches with the help of several illuminative (counter)examples. Furthermore, we discuss marginal standardization, which is a useful tool in classical multivariate extreme value theory but, as we point out, much less straightforward and sometimes even obscuring in conditional extreme value models. Finally, we indicate how, in some situations, a more comprehensive characterization of the asymptotic behavior can be obtained if the conditions of conditional extreme value models are relaxed so that the limit is no longer unique.     

Vector generalized linear and additive extreme value models

Over recent years parametric and nonparametric regression has slowly been adopted into extreme value data analysis. Its introduction has been characterized by piecemeal additions and embellishments, which has had a negative effect on software development and usage. The purpose of this article is to convey the classes of vector generalized linear and additive models (VGLMs and VGAMs) as offering significant advantages for extreme value data analysis, providing flexible smoothing within a unifying framework. In particular, VGLMs and VGAMs allow all parameters of extreme value distributions to be modelled as linear or smooth functions of covariates. We implement new auxiliary methodology by incorporating a quasi-Newton update for the working weight matrices within an iteratively reweighted least squares (IRLS) algorithm. A software implementation by the first author, called the vgam package for , is used to illustrate the potential of VGLMs and VGAMs.     

分组数据下线性回归模型的参数估计

讨论了分组数据下线性回归模型参数的MLE的存在、唯一性.通过EM算法获得MLE的近似解.通过SEM算法获得MLE的渐近协方差阵.     

Comparing extreme models when the sign of the extreme value index is known

In the literature on analyzing extremes, both generalized Pareto distributions and Pareto distributions are employed to infer the tail of a distribution with a known positive extreme value index. Similar studies exist for a known negative extreme value index. Intuitively, one should not employ the generalized Pareto distribution in the case of knowing the sign of the extreme value index. In this work, we show that fitting a generalized Pareto distribution is equivalent to the model in Hall (1982) in the case of a negative extreme value index, in both improving the rate of convergence and including the bias term of the asymptotic results of that reference. When the extreme value index is known to be positive, we show that fitting a generalized Pareto distribution may be preferred in some cases determined by a so-called second-order parameter and the extreme value index itself.     

A choice probability characterization of generalized extreme value models

A set of necessary and sufficient conditions is established for the representability of choice probabilities by additive random utility models with generalized extreme value (GEV) distributions of utilities. These conditions yield an operational testing procedure for GEV-representability which does not require explicit construction of the underlying distribution of utilities. In addition, this characterization of GEV models reveals a number of their underlying behavioral features.     

Averaged extreme regression quantile

Various events in the nature, economics and in other areas force us to combine the study of extremes with regression and other methods. A useful tool for reducing the role of nuisance regression, while we are interested in the shape or tails of the basic distribution, is provided by the averaged regression quantile and namely by the average extreme regression quantile. Both are weighted means of regression quantile components, with weights depending on the regressors. Our primary interest is the averaged extreme regression quantile (AERQ), its structure, qualities and its applications, e.g. in investigation of a conditional loss given a value exogenous economic and market variables. AERQ has several interesting equivalent forms: While it is originally defined as an optimal solution of a specific linear programming problem, hence is a weighted mean of responses corresponding to the optimal base of the pertaining linear program, we give another equivalent form as a maximum residual of responses from a specific R-estimator of the slope components of regression parameter. The latter form shows that while AERQ equals to the maximum of some residuals of the responses, it has minimal possible perturbation by the regressors. Notice that these finite-sample results are true even for non-identically distributed model errors, e.g. under heteroscedasticity. Moreover, the representations are formally true even when the errors are dependent - this all provokes a question of the right interpretation and of other possible applications.     

Hurdle models of loan default

Some models of loan default are binary, simply modelling the probability of default, while others go further and model the extent of default (eg number of outstanding payments; amount of arrears). The double-hurdle model, originally due to Cragg (Econometrica, 1971), and conventionally applied to household consumption or labour supply decisions, contains two equations, one which determines whether or not a customer is a potential defaulter (the ‘first hurdle’), and the other which determines the extent of default. In separating these two processes, the model recognizes that there exists a subset of the observed non-defaulters who would never default whatever their circumstances. A Box-Cox transformation applied to the dependent variable is a useful generalization to the model. Estimation is relatively easy using the Maximum Likelihood routine available in STATA. The model is applied to a sample of 2515 loan applicants for whom loans were approved, a sizeable proportion of whom defaulted in varying degrees. The dependent variables used are amount in arrears and number of days in arrears. The value of the hurdle approach is confirmed by finding that certain key explanatory variables have very different effects between the two equations. Most notably, the effect of loan amount is strongly positive on arrears, while being U-shaped on the probability of default. The former effect is seriously under-estimated when the first hurdle is ignored.     

Estimating extreme bivariate quantile regions

When simultaneously monitoring two possibly dependent, positive risks one is often interested in quantile regions with very small probability p. These extreme quantile regions contain hardly any or no data and therefore statistical inference is difficult. In particular when we want to protect ourselves against a calamity that has not yet occurred, we need to deal with probabilities p?<?1/n, with n the sample size. We consider quantile regions of the form {(x, y)?∈?(0, ∞?)²: f(x, y)?≤?β}, where f, the joint density, is decreasing in both coordinates. Such a region has the property that it consists of the less likely points and hence that its complement is as small as possible. Using extreme value theory, we construct a natural, semiparametric estimator of such a quantile region and prove a refined form of consistency. A detailed simulation study shows the very good statistical performance of the estimated quantile regions. We also apply the method to find extreme risk regions for bivariate insurance claims.     

Support vector regression for loss given default modelling

Loss given default modelling has become crucially important for banks due to the requirement that they comply with the Basel Accords and to their internal computations of economic capital. In this paper, support vector regression (SVR) techniques are applied to predict loss given default of corporate bonds, where improvements are proposed to increase prediction accuracy by modifying the SVR algorithm to account for heterogeneity of bond seniorities. We compare the predictions from SVR techniques with thirteen other algorithms. Our paper has three important results. First, at an aggregated level, the proposed improved versions of support vector regression techniques outperform other methods significantly. Second, at a segmented level, by bond seniority, least square support vector regression demonstrates significantly better predictive abilities compared with the other statistical models. Third, standard transformations of loss given default do not improve prediction accuracy. Overall our empirical results show that support vector regression techniques are a promising technique for banks to use to predict loss given default.     

Comparison of location models of Weibull type samples and extreme value processes

Summary Four different location parameter models are compared within the sufficiency and deficiency concept. The starting is a location model of a Weibull type sample with shape parameter -1<a<1. Here our basic inequality concerns the approximate sufficiency of the k lower extremes. In addition, the lower extremes are approximately equal, in distribution, to where S _m is the sum of m i.i.d. standard exponential random variables and t is the location parameter. The final step leads us to the model of extreme value processes ...     

Fisher information for generalised linear mixed models

The Fisher information for the canonical link exponential family generalised linear mixed model is derived. The contribution from the fixed effects parameters is shown to have a particularly simple form.     

Combined regression models

The goal of this paper is two-fold. First, new regression models obtained by combinations of the least squares (LS), minimax (MM), and the least sum of absolute deviations (LSAD) are proposed. Second, measures for assessing the influence of observations on the fitted models are suggested. The paper is interdisciplinary because the theory behind the proposed method draws from results in the operations research area. The methods are illustrated by their application to some examples and graphical illustrations are given.     

Two-step generalised empirical likelihood inference for semiparametric models

This paper shows how the generalised empirical likelihood method can be used to obtain valid asymptotic inference for the finite dimensional component of semiparametric models defined by a set of moment conditions. The results of the paper are illustrated using three well-known semiparametric regression models: partially linear single index, linear transformation with random censoring, and quantile regression with random censoring. Monte Carlo simulations suggest that some of the proposed test statistics have competitive finite sample properties. The results of the paper are applied to test for functional misspecification in a hedonic price model of a housing market.     

Estimation of the extreme value and the extreme points

Summary Letf be a continuous function defined on some domainA andX ₁,X ₂, ... be iid random variables. We estimate the extreme value off onA by studying the limiting distribution of min {f(X ₁), ...,f(X _n )} or max {f(X ₁), ...,f(X _n )} properly normalized. Sufficient conditions for the existence of the limiting distribution as well as a characterization of the limiting distribution relative to the extreme points off will be provided. A discussion of the multidimensional case is also carried out. Partially supported by CNPq-No. 301508/84.     

On testing extreme value conditions

Applications of univariate extreme value theory rely on certain as- sumptions. Recently, two methods for testing these extreme value conditions are derived by [Dietrich, D., de Haan, L., Hüsler, J., Extremes 5: 71–85, (2002)] and [Drees, H., de Haan, L., Li, D., J. Stat. Plan. Inference, 136: 3498–3538, (2006)]. In this paper we compare the two tests by simulations and investigate the effect of a possible weight function by choosing a parameter, the test error and the power of each test. The conclusions are useful for extreme value applications.