We study the impact of wrong-way risk (WWR) on the credit valuation adjustment (CVA) of a portfolio of interest rate swaps (IRSs), using an intensity-based reduced form model. To model WWR in IRSs we create a dependence between he underlying market risk factor of the IRS and the survival probability of the
counterparty. The focus lies on modeling the credit part of the CVA, choosing the most suitable default model. Using a Monte Carlo (MC) framework, we correlate the Brownian increments of the stochastic processes that describe the interest rate and the hazard rate. Assuming the correlation parameter is chosen based on historical data, we solve the problem of leaking correlation. Two
stochastic processes for the credit part are chosen, the CIR++ and the JCIR++, such that the influence of the model choice can be separated from the impact of WWR. With a case study, we vary the correlation parameter to quantify the impact of WWR on CVA of the portfolio. We show that the introduced dependence has a significant impact on the CVA, that depends on the calibration, the portfolio and the chosen correlation. Solving the leaking correlation problem can increase the impact of WWR on the CVA, especially using JCIR++ dynamics.

Phylogenetic networks represent the evolutionary history of organisms and the relationships among them. In the field of phylogenetics research has been done to reconstruct these networks. Several methods for reconstructing those networks have been developed over the years. One of them is the softwired parsimony score and another, a fairly new method, is the parental parsimony method. In theory this is a better and more accurate method, but before this thesis, it was not tested yet. In this thesis, we compare the softwired and the parental parsimony method in practice by implementing the methods and calculating their parsimony scores. The results will show that the parental parsimony is a better and more accurate method in practice as well.