GAUK No. 1278218 Credit Transition matrices based on bank-sourced data and the business cycle
|Řešitel:||Ing. Mgr. Barbora Štěpánková (Máková) M.A.|
prof. PhDr. Ladislav Krištoufek Ph.D.
|Popis:||The research will focus on estimation of credit transition matrices (CTMs) using a unique dataset of probabilities of default received from global banks provided by Credit Benchmark. CTMs describe transitions between different credit rating categories over a period of time. They are important inputs to many risk management applications such as bond and credit derivatives pricing and the regulatory capital assessment. The latest CECL and IFRS9 accounting rules require banks to estimate potential losses over the entire life of a loan and CTMs can provide an insight into the likely pattern of losses over various time horizons.
Currently the main sources of CTMs are credit rating agencies (CRAs). The bank sourced data available to me can be used for constructing alternative CTMs representing market view. My research goal is to describe characteristics of bank-sourced credit risk data and choose an appropriate model to estimate the CTMs. I will test their stability and compare them to CTMs published by CRAs; further I will construct robust industry and region-specific matrices. I will also consider the difference between through-the-cycle and point-in-time probabilities of default and effect of these characteristics on Markov process and time-homogeneity tests and discuss possibilities of creating point-in-time CTM based on through-the-cycle CTM, macroeconomic variables and market data. Finally, I will use the CTMs to calculate capital requirements for an imaginary portfolio.
|Práce v rámci grantu:|
|Finance:||GAUK No. 1278218|