Essays on the term structure of interest rates

This PhD thesis contains three main chapters on macro finance, with a focus on the term structure of interest rates and the applications of state-of-the-art Bayesian econometrics. Except for Chapter 1 and Chapter 5, which set out the general introduction and conclusion, each of the chapters can be considered as a standalone piece of work. In Chapter 2, we model and predict the term structure of US interest rates in a data rich environment. We allow the model dimension and parameters to change over time, accounting for model uncertainty and sudden structural changes. The proposed timevarying parameter Nelson-Siegel Dynamic Model Averaging (DMA) predicts yields better than standard benchmarks. DMA performs better since it incorporates more macro-finance information during recessions. The proposed method allows us to estimate plausible realtime term premia, whose countercyclicality weakened during the financial crisis. Chapter 3 investigates global term structure dynamics using a Bayesian hierarchical factor model augmented with macroeconomic fundamentals. More than half of the variation in the bond yields of seven advanced economies is due to global co-movement. Our results suggest that global inflation is the most important factor among global macro fundamentals. Non-fundamental factors are essential in driving global co-movements, and are closely related to sentiment and economic uncertainty. Lastly, we analyze asymmetric spillovers in global bond markets connected to diverging monetary policies. Chapter 4 proposes a no-arbitrage framework of term structure modeling with learning and model uncertainty. The representative agent considers parameter instability, as well as the uncertainty in learning speed and model restrictions. The empirical evidence shows that apart from observational variance, parameter instability is the dominant source of predictive variance when compared with uncertainty in learning speed or model restrictions. When accounting for ambiguity aversion, the out-of-sample predictability of excess returns implied by the learning model can be translated into significant and consistent economic gains over the Expectations Hypothesis benchmark.

Essays on the term structure of interest rates

This PhD thesis contains three main chapters on macro finance, with a focus on the term structure of interest rates and the applications of state-of-the-art Bayesian econometrics. Except for Chapter 1 and Chapter 5, which set out the general introduction and conclusion, each of the chapters can be considered as a standalone piece of work. In Chapter 2, we model and predict the term structure of US interest rates in a data rich environment. We allow the model dimension and parameters to change over time, accounting for model uncertainty and sudden structural changes. The proposed time-varying parameter Nelson-Siegel Dynamic Model Averaging (DMA) predicts yields better than standard benchmarks. DMA performs better since it incorporates more macro-finance information during recessions. The proposed method allows us to estimate plausible real-time term premia, whose countercyclicality weakened during the financial crisis. Chapter 3 investigates global term structure dynamics using a Bayesian hierarchical factor model augmented with macroeconomic fundamentals. More than half of the variation in the bond yields of seven advanced economies is due to global co-movement. Our results suggest that global inflation is the most important factor among global macro fundamentals. Non-fundamental factors are essential in driving global co-movements, and are closely related to sentiment and economic uncertainty. Lastly, we analyze asymmetric spillovers in global bond markets connected to diverging monetary policies. Chapter 4 proposes a no-arbitrage framework of term structure modeling with learning and model uncertainty. The representative agent considers parameter instability, as well as the uncertainty in learning speed and model restrictions. The empirical evidence shows that apart from observational variance, parameter instability is the dominant source of predictive variance when compared with uncertainty in learning speed or model restrictions. When accounting for ambiguity aversion, the out-of-sample predictability of excess returns implied by the learning model can be translated into significant and consistent economic gains over the Expectations Hypothesis benchmark.

Modelling the development of an online learning resource by health care professionals

The aim of this study was to model the process of development for an Online Learning Resource (OLR) by Health Care Professionals (HCPs) to meet lymphoedema-related educational needs, within an asset-based management context. Previous research has shown that HCPs have unmet educational needs in relation to lymphoedema but details on their specific nature or context were lacking. Against this background, the study was conducted in two distinct but complementary phases. In Phase 1, a national survey was conducted of HCPs predominantly in community, oncology and palliative care services, followed by focus group discussions with a sample of respondents. In Phase 2, lymphoedema specialists (LSs) used an action research approach to design and implement an OLR to meet the needs identified in Phase 1. Study findings were analysed using descriptive statistics (Phase 1), and framework, thematic and dialectic analysis to explore their potential to inform future service development and education theory. Unmet educational need was found to be specific to health care setting and professional group. These resulted in HCPs feeling poorly-equipped to diagnose and manage lymphoedema. Of concern, when identified, lymphoedema was sometimes buried for fear of overwhelming stretched services. An OLR was identified as a means of addressing the unmet educational needs. This was successfully developed and implemented with minimal additional resources. The process model created has the potential to inform contemporary leadership theory in asset-based management contexts. This doctoral research makes a timely contribution to leadership theory since the resource constraints underpinning much of the contribution has salience to current public services. The process model created has the potential to inform contemporary leadership theory in asset-based management contexts. Further study of a leadership style which incorporates cognisance of Cognitive Load Theory and Self-Determination Theory is suggested. In addition, the detailed reporting of process and how this facilitated learning for participants contributes to workplace education theory

Using interaction data for improving the offline and online evaluation of search engines

This thesis investigates how web search evaluation can be improved using historical interaction data. Modern search engines combine offline and online evaluation approaches in a sequence of steps that a tested change needs to pass through to be accepted as an improvement and subsequently deployed. We refer to such a sequence of steps as an evaluation pipeline. In this thesis, we consider the evaluation pipeline to contain three sequential steps: an offline evaluation step, an online evaluation scheduling step, and an online evaluation step. In this thesis we show that historical user interaction data can aid in improving the accuracy or efficiency of each of the steps of the web search evaluation pipeline. As a result of these improvements, the overall efficiency of the entire evaluation pipeline is increased. Firstly, we investigate how user interaction data can be used to build accurate offline evaluation methods for query auto-completion mechanisms. We propose a family of offline evaluation metrics for query auto-completion that represents the effort the user has to spend in order to submit their query. The parameters of our proposed metrics are trained against a set of user interactions recorded in the search engine’s query logs. From our experimental study, we observe that our proposed metrics are significantly more correlated with an online user satisfaction indicator than the metrics proposed in the existing literature. Hence, fewer changes will pass the offline evaluation step to be rejected after the online evaluation step. As a result, this would allow us to achieve a higher efficiency of the entire evaluation pipeline. Secondly, we state the problem of the optimised scheduling of online experiments. We tackle this problem by considering a greedy scheduler that prioritises the evaluation queue according to the predicted likelihood of success of a particular experiment. This predictor is trained on a set of online experiments, and uses a diverse set of features to represent an online experiment. Our study demonstrates that a higher number of successful experiments per unit of time can be achieved by deploying such a scheduler on the second step of the evaluation pipeline. Consequently, we argue that the efficiency of the evaluation pipeline can be increased. Next, to improve the efficiency of the online evaluation step, we propose the Generalised Team Draft interleaving framework. Generalised Team Draft considers both the interleaving policy (how often a particular combination of results is shown) and click scoring (how important each click is) as parameters in a data-driven optimisation of the interleaving sensitivity. Further, Generalised Team Draft is applicable beyond domains with a list-based representation of results, i.e. in domains with a grid-based representation, such as image search. Our study using datasets of interleaving experiments performed both in document and image search domains demonstrates that Generalised Team Draft achieves the highest sensitivity. A higher sensitivity indicates that the interleaving experiments can be deployed for a shorter period of time or use a smaller sample of users. Importantly, Generalised Team Draft optimises the interleaving parameters w.r.t. historical interaction data recorded in the interleaving experiments. Finally, we propose to apply the sequential testing methods to reduce the mean deployment time for the interleaving experiments. We adapt two sequential tests for the interleaving experimentation. We demonstrate that one can achieve a significant decrease in experiment duration by using such sequential testing methods. The highest efficiency is achieved by the sequential tests that adjust their stopping thresholds using historical interaction data recorded in diagnostic experiments. Our further experimental study demonstrates that cumulative gains in the online experimentation efficiency can be achieved by combining the interleaving sensitivity optimisation approaches, including Generalised Team Draft, and the sequential testing approaches. Overall, the central contributions of this thesis are the proposed approaches to improve the accuracy or efficiency of the steps of the evaluation pipeline: the offline evaluation frameworks for the query auto-completion, an approach for the optimised scheduling of online experiments, a general framework for the efficient online interleaving evaluation, and a sequential testing approach for the online search evaluation. The experiments in this thesis are based on massive real-life datasets obtained from Yandex, a leading commercial search engine. These experiments demonstrate the potential of the proposed approaches to improve the efficiency of the evaluation pipeline.