Wolfgang Karl Härdle

A Confidence Corridor for Expectile Functions

Let (X1; Y1), ..., (Xn; Yn) be i.i.d. rvs and let v(x) be the unknown tau - expectile regression curve of Y conditional on X. An expectile-smoother vn(x) is a localized, nonlinear estimator of v(x). The strong uniform consistency rate is established under general conditions. In many applications it is necessary to know the stochastic fluctuation of the process {vn(x)-v(x)}. Using strong approximations of the empirical process and extreme value theory, we consider the asymptotic maximal deviation sup06x61 jvn(x)o€€€v(x)j. The derived result helps in the construction of a uniform confidence band for the expectile curve v(x). This paper considers fitting a simultaneous confidence corridor (SCC) around the estimated expectile function of the conditional distribution of Y given x based on the observational data generated according to a nonparametric regression model. Moreover, we construct the simultaneous confidence corridors around the expectiles of the residuals from the temperature models to investigate the temperature risk drivers.

Quantile Regression in Risk Calibration

Financial risk control has always been challenging and becomes now an even harder problem as joint extreme events occur more frequently. For decision makers and government regulators, it is therefore important to obtain accurate information on the interdependency of risk factors. Given a stressful situation for one market participant, one likes to measure how this stress affects other factors. The CoVaR (Conditional VaR) framework has been developed for this purpose. The basic technical elements of CoVaR estimation are two levels of quantile regression: one on market risk factors; another on individual risk factor. Tests on the functional form of the two-level quantile regression reject the linearity. A flexible semiparametric modeling framework for CoVaR is proposed. A partial linear model (PLM) is analyzed. In applying the technology to stock data covering the crisis period, the PLM outperforms in the crisis time, with the justification of the backtesting procedures. Moreover, using the data on global stock markets indices, the analysis on marginal contribution of risk (MCR) defined as the local first order derivative of the quantile curve sheds some light on the source of the global market risk.

Composite Quantile Regression for the Single-Index Model

Quantile regression is in the focus of many estimation techniques and is an important tool in data analysis. When it comes to nonparametric specifications of the conditional quantile (or more generally tail) curve one faces, as in mean regression, a dimensionality problem. We propose a projection based single index model specifi- cation. For very high dimensional regressors X one faces yet another dimensionality problem and needs to balance precision vs. dimension. Such a balance may be achieved by combining semiparametric ideas with variable selection techniques.

Pricing Chinese rain: a multi-site multi-period equilibrium pricing model for rainfall derivatives

Many industries are exposed to weather risk which they can transfer on financial markets via weather derivatives. Equilibrium models based on partial market clearing became a useful tool for pricing such kind of financial instruments. In a multi-period equilibrium pricing model agents rebalance their portfolio of weather bonds and a risk free asset in each period such that they maximize the expected utility of their incomes constituted by possibly weather dependent profits and payoffs of portfolio positions. We extend the model to a multisite version and apply it to pricing rainfall derivatives for Chinese provinces. By simulating realistic market conditions with two agent types, farmers with profits highly exposed to weather risk and a financial investor diversifying her financial portfolio, we obtain equilibrium prices for weather derivatives on cumulative monthly rainfall. Dynamic portfolio optimization under market clearing and utility indifference of these representative agents determines equilibrium quantity and price for rainfall derivatives.

Learning Machines Supporting Bankruptcy Prediction

In many economic applications it is desirable to make future predictions about the financial status of a company. The focus of predictions is mainly if a company will default or not. A support vector machine (SVM) is one learning method which uses historical data to establish a classification rule called a score or an SVM. Companies with scores above zero belong to one group and the rest to another group. Estimation of the probability of default (PD) values can be calculated from the scores provided by an SVM. The transformation used in this paper is a combination of weighting ranks and of smoothing the results using the PAV algorithm. The conversion is then monotone. This discussion paper is based on the Creditreform database from 1997 to 2002. The indicator variables were converted to financial ratios; it transpired out that eight of the 25 were useful for the training of the SVM. The results showed that those ratios belong to activity, profitability, liquidity and leverage. Finally, we conclude that SVMs are capable of extracting the necessary information from financial balance sheets and then to predict the future solvency or insolvent of a company. Banks in particular will benefit from these results by allowing them to be more aware of their risk when lending money.

Reference Dependent Preferences and the EPK Puzzle

Supported by several recent investigations, the empirical pricing kernel (EPK) puzzle might be considered a stylized fact. Based on an economic model with state dependent preferences for the financial investors, we want to emphasize a microeconomic view that succeeds in explaining the puzzle. We retain the expected utility framework in a one period model and illustrate the case when the state is defined with respect to a reference point. We further investigate how the model relates the shape of the EPK to the economic conditions.

Statistics E-Learning Platforms Evaluation: Case Study

With the increase of e-learning by universities and educational institutes in the world through more electronic platforms, come the questions to researchers, educators and designers of electronic platforms about feasibility and using this method of learning. Are we achieving the desired goals and improving the quality of education? Are we improving their performance and ability to self-study without the need for a teacher? Is e-learning an effective and successful method from the students views? In this paper, we consider evaluate e-learning systems in statistics. We make an evaluation study, we analyze a students sample of the methods: Factor analysis, Logit model. The common aim of this evaluation is to provide data to justify the results or evidence to support that the e-learning platforms are helping the students to learn more effectively. The questionnaire covers information about e-learning evaluation criterias. Some of these criterias are: Navigability, applicability, instructional structure and interactivity.