CEU Electronic Theses and Dissertations, 2016
| Author | Tran, Dávid Manh Ha |
|---|---|
| Title | An Optimal Debt Management Strategy for Hungary |
| Summary | This thesis analyzes the expected costs and risks associated with Hungarian government securities with the purpose of constructing an optimal debt management strategy over a five-year horizon. In order to accomplish this objective, two intermediate issues are resolved. First, different empirical yield curve models are compared based on in-sample fitting. I find that more complex models fit the yield curve better, but additional restrictions are necessary for them to be interpretable as factor models. Second, the yield curve models are compared based on pseudo out-of-sample forecasting performance. Using various autoregressive, vector-autoregressive and state-space models for prediction, I conclude that the Nelson-Siegel model in a state-space framework outperforms the benchmark random walk forecast on longer maturities and horizons. This model is then used to carry out an ex ante prediction for the 2016-2020 horizon. Ten thousand scenarios are generated using a Monte-Carlo simulation describing the evolution of the term structure of interest rates. The scenarios are used to calculate the expected debt charges associated with bonds. Finally, the mean-variance and the mean-conditional cost-at-risk efficient frontiers are established. The policy conclusion is that reducing the share of very long and very short maturity government bonds in favor of medium maturity papers would be a Pareto-improvement to the current Hungarian debt portfolio. |
| Supervisor | Bokros, Lajos András |
| Department | Economics MA |
| Full text | https://www.etd.ceu.edu/2016/tran_david.pdf |
Visit the CEU Library.
© 2007-2021, Central European University