Research Project 4

RP4. Variable Annuity ALM (Philipp Ustinov)

Leading Partner: CSA. Supervisor: Prof. Michael Dempster, Managing Director

This project will involve familiarization with current CSA software and its application to the development of a full dynamic stochastic programming (SDP) solution to the asset liability management of a large scale portfolio of issued variable annuities. In contrast to a derivative based LDI hedging solution as developed in WP1, this approach makes a full cash flow match of assets (at the asset class level) and liabilities over a long term planning horizon. This approach involves developing econometric asset and liability forecasting models, variance reduced Monte Carlo forward simulation of these models and large scale optimisation of the ALM model using decomposition methods tuned to the HPC techniques and technologies developed inWP5 (RP 11-13). The ER who will be supported by CSA personnel will develop all these skills in the project.

Results

Deliverables

DELIVERABLE 2.3 on project 4 (P. Ustinov, CSA): Dempster, M.A.H., Medova, E.A., Osmolovskiy, I. & Ustinov*, P. (2015), "Developing a practical robust long term yield curve model" [download]

Publications

Dempster, M.A.H., Kloppers, D., Medova, E., Osmolovskiy, I. & Ustinov, P. (2016), “Life cycle goal achievement or portfolio volatility reduction?”. Journal of Portfolio Management, Vol. 42, No.2, pp. 99-117. [download]


Final conference

Philipp Ustinov (CSA): Yield curve modelling in a low rate environment