Proofs for the consistency of the kernel density estimator have historically developed. Four important milestones are the pointwise consistency, the almost sure uniform convergence, the rate of convergence on a bounded interval and the rate of convergence on R. The underlying concepts of total...

Multivariate equivalence testing becomes necessary whenever the similarity rather than a difference between several treatment groups with multiple endpoints has to be shown. This problem occurs in various applications, including bioequivalence or the comparison of dissolution profiles....

We consider the problem of uniform asymptotics in kernel functional estimation where the bandwidth can depend on the data. In a unified approach we investigate kernel estimates of the density and the hazard rate for uncensored and right-censored observations. The model allows for the fixed...

When calculating the cost of entering into a credit transaction the predominant stochastic component is the expected loss. Often in the credit business the one-year probability of default of the liable counterpart is the only reliable parameter. We use this probability to calculating the exact...

We derive approximate formulae for the credit value-at-risk and the economic capital of a large credit portfolio. The representation allows to change the risk horizon quickly and avoids simulation or numerical procedures. The Poisson mixture model is equivalent to CreditRisk and uses the same...

Most credit portfolio models exclusively calculate the loss distribution for a portfolio of performing counterparts. Conservative default definitions cause considerable insecurity about the loss for a long time after the default. We present three approaches to account for defaulted counterparts...