Robust optimization models for managing callable bond portfolios

Article Abstract:

Two optimization models for managing portfolios of callable bonds have been developed. The single-period model ensures robustness by penalizing downside error. The other model is a multi-stage stochastic program with recourse. Backtests on both models using ex poste market data show that they perform consistently well. The stochastic portfolio achieved a greater annual overperformance compared to the robust optimization portfolio.

Research, Bonds, Bonds (Securities)

---

A stochastic programming model for money management

Article Abstract:

Portfolio managers need innovative techniques to assess fixed-income markets such as mortgages, derivatives and high-yield bonds. The money management strategy presented in the study is a multi-period, stochastic programming problem with recourse. The model can be hedged against various forms of uncertainties in fixed-income markets and can provide superior results compared to dynamic, single-period stochastic models.

Financial planning

---

Similar abstracts:

• Abstracts: Modelling and optimisation for sustainable development policy assessment. On the distribution of calls in an wireless network driven by fluid traffic
• Abstracts: Reducing simulation models for scheduling manufacturing facilities. A contribution to the stochastic flow shop scheduling problem
• Abstracts: Economic order quantity models for ameliorating/deteriorating items under inflation and time discounting. Providing support for decisions based on time series information under conditions of asymmetric loss
• Abstracts: Application of multi-steps forecasting for restraining the bullwhip effect and improving inventory performance under autoregressive demand
• Abstracts: Static hedging of multivariate derivatives by simulation. Estimation of asset demands by heterogeneous agents

This website is not affiliated with document authors or copyright owners. This page is provided for informational purposes only. Unintentional errors are possible.