Courtesy of the publishers of the papers underlying chapters 2 and 3, the thesis is available through the WebDOC system of the Econometric Institute as a PDF document (1.5MB).
| Acknowledgements | |
|---|---|
| 1 | Introduction |
| 1.1 | Motivation |
| 1.2 | Inflation: Long memory |
| 1.3 | Exchange rate: Varying trend and variance |
| 1.4 | Overview |
| 2 | Long Memory and Level Shifts: Re-Analysing Inflation Rates |
| 2.1 | Introduction |
| 2.2 | A motivation |
| 2.3 | Some theoretical results |
| 2.4 | Simulation evidence |
| 2.5 | Inflation: Long memory and level shifts |
| 2.6 | Conclusions |
| 2.A | Calculating the likelihood and the test statistics |
| 2.B | Generating an ARFIMA process |
| 2.C | Data sources, programs, justifications |
| 3 | Inflation, Forecast Intervals and Long Memory Regression Models |
| 3.1 | Introduction |
| 3.2 | Recursive ARFIMAX forecasting |
| 3.2.1 | Basic features of U.S. core inflation |
| 3.2.2 | ARFIMAX modelling |
| 3.2.3 | Recursive estimation and forecasting |
| 3.2.4 | Recursive estimates |
| 3.2.5 | Recursive forecasting |
| 3.3 | Recursive weighted ARFIMAX forecasting |
| 3.4 | Recursive ARFIMAX forecast tests |
| 3.5 | Conclusion |
| 4 | Bayesian Sampling Methods |
| 4.1 | Introduction |
| 4.2 | Basic sampling methods |
| 4.2.1 | Importance sampling |
| 4.2.2 | Sampling/Importance Resampling |
| 4.2.3 | Acceptance-rejection sampling |
| 4.2.4 | The Metropolis-Hastings algorithm |
| 4.2.5 | Gibbs sampling |
| 4.2.6 | On convergence: Theory |
| 4.2.7 | On convergence: Practice |
| 4.3 | Extensions |
| 4.3.1 | Metropolis-within-Gibbs |
| 4.3.2 | Multistep Gibbs samplers revisited |
| 4.3.3 | Griddy Gibbs sampler |
| 4.3.4 | Adaptive Polar Sampling |
| 4.4 | Posterior odds and marginal likelihood |
| 4.4.1 | Posterior odds, the Bayes factor, and model choice |
| 4.4.2 | Calculating the marginal likelihood |
| 4.4.3 | Calculating the Bayes factor |
| 4.5 | Example: Sampling from a time series model |
| 4.5.1 | Introducing the model |
| 4.5.2 | Likelihood and posterior |
| 4.5.3 | Sampling methods based on the likelihood function |
| 4.5.4 | Sampling methods based on the conditional densities |
| 4.5.5 | Calculating the marginal likelihood of the model |
| 4.5.6 | Sampling setup and details |
| 4.6 | Concluding remarks |
| 4.A | Sampling and the state space model |
| 4.B | Derivations and distributions for APS |
| 4.C | Selected density functions |
| 5 | Daily Hedging of Currency Risk |
| 5.1 | Introduction |
| 5.2 | Currency hedging |
| 5.3 | Time series models for exchange rate returns |
| 5.4 | Bayesian inference and decision |
| 5.4.1 | Prior structure |
| 5.4.2 | Constructing a posterior sample |
| 5.4.3 | Evaluating the marginal likelihood |
| 5.4.4 | Predictive analysis |
| 5.4.5 | Decision analysis |
| 5.5 | Exchange rate data and the interest rate |
| 5.5.1 | Stylized facts |
| 5.5.2 | Data in the evaluation period |
| 5.6 | Convergence of MCMC and posterior results |
| 5.6.1 | The posterior distribution |
| 5.6.2 | Marginal likelihood |
| 5.6.3 | Predictive density of the models |
| 5.7 | Hedging results |
| 5.7.1 | On the setup |
| 5.7.2 | Naive hedging strategies |
| 5.7.3 | Variability of the hedging decision |
| 5.7.4 | Returns and utilities of model-based strategies |
| 5.7.5 | Value-at-Risk and the Sharpe ratio |
| 5.7.6 | Other viewpoints on the results |
| 5.8 | Concluding remarks |
| 5.A | Gibbs sampling with data augmentation |
| 6 | Conclusions |
| 6.1 | General remarks |
| 6.2 | Inflation rates |
| 6.3 | Bayesian simulation methods |
| 6.4 | Hedging currency risk |
| Bibliography | |
| Nederlandse samenvatting |