In dandermotj/basel: Implementation of the Basel III Risk Weight Functions

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>"
)

Internal Ratings Based Approach Models

IRB models can produce two outputs. In the case of Foundation IRB that output is a probability of default (PD). For Advanced IRB that is a probability of default and a loss given default (LGD). Using the PD and the LGD we can compute the capital requirement for a borrower. We can then compute the risk weighted assets of that borrower.

Steps to computing tier 1 capital:

1. Compute 1-year PD
2. Compute capital requirement
3. Compute total RWA
4. Compute Tier 1 capital

These are actually quite simple steps, but amazingly this process can take up to a year to complete.

Computing Capital

So now we have a dataset of borrowers, their exposure at each point in time and a PD. Given we are doing a foundation IRB model, we will use the same LGD for every borrower, prescribed by Basel as X.

lending_club %>% 
  mutate(K = capital_requirement(PD, LGD = ))

dandermotj/basel documentation built on May 21, 2019, 10:10 a.m.