P&L Attribution Framework using the revaluation method
Objective
- The client wanted to implement a framework to perform P&L attribution using the revaluation method as it is more accurate
CRISIL's solution
- A user can trigger attribution computation at the launch of a command. The relevant risk is configurable as input parameters to the command - A daemon process that runs in the background and responds to the user trigger
- Based on the user's inputs, the process creates market data scenarios in the database and revaluates the book based on newly generated scenarios
- Each revaluation generates an output file, which is picked up by a front-end application called the risk viewer, displaying the numbers in suitable formats
- Used technologies such as C++ and Shell scripting
Client impact
- The CRISIL solution enabled P&L attribution w.r.t each risk parameter at the launch of a command through customisable and intuitive input parameters
- Identify exhaustive list of categories for P&L movement
- Facilitate preparation of a meaningful commentary on the PL changes and spot any PL change beyond a meaningful level
- Developed control function of the bank to work independently and explain daily PL movement
Questions