Problems and Algorithms for Sequence Segmentations

The analysis of sequential data is required in many diverse areas such as telecommunications, stock market analysis, and bioinformatics. A basic problem related to the analysis of sequential data is the sequence segmentation problem. A sequence segmentation is a partition of the sequence into a number of non-overlapping segments that cover all data points, such that each segment is as homogeneous as possible. This problem can be solved optimally using a standard dynamic programming algorithm. In the first part of the thesis, we present a new approximation algorithm for the sequence segmentation problem. This algorithm has smaller running time than the optimal dynamic programming algorithm, while it has bounded approximation ratio. The basic idea is to divide the input sequence into subsequences, solve the problem optimally in each subsequence, and then appropriately combine the solutions to the subproblems into one final solution. In the second part of the thesis, we study alternative segmentation models that are devised to better fit the data. More specifically, we focus on clustered segmentations and segmentations with rearrangements. While in the standard segmentation of a multidimensional sequence all dimensions share the same segment boundaries, in a clustered segmentation the multidimensional sequence is segmented in such a way that dimensions are allowed to form clusters. Each cluster of dimensions is then segmented separately. We formally define the problem of clustered segmentations and we experimentally show that segmenting sequences using this segmentation model, leads to solutions with smaller error for the same model cost. Segmentation with rearrangements is a novel variation to the segmentation problem: in addition to partitioning the sequence we also seek to apply a limited amount of reordering, so that the overall representation error is minimized. We formulate the problem of segmentation with rearrangements and we show that it is an NP-hard problem to solve or even to approximate. We devise effective algorithms for the proposed problem, combining ideas from dynamic programming and outlier detection algorithms in sequences. In the final part of the thesis, we discuss the problem of aggregating results of segmentation algorithms on the same set of data points. In this case, we are interested in producing a partitioning of the data that agrees as much as possible with the input partitions. We show that this problem can be solved optimally in polynomial time using dynamic programming. Furthermore, we show that not all data points are candidates for segment boundaries in the optimal solution.

Technology-related peso problems in stock returns

Although empirical evidence suggests the contrary, many asset pricing models assume stock returns to be symmetrically distributed. In this paper it is argued that the occurrence of negative jumps in a firm's future earnings and, consequently, in its stock price, is positively related to the level of network externalities in the firm's product market. If the ex post frequency of these negative jumps in a sample does not equal the ex ante assessed probability of occurrence, the sample is subject to a peso problem. The hypothesis is tested for by regressing the skewness coefficient of a firm’s realised stock return distribution on the firm’s R&D intensity, i.e. the ratio of the firm’s research and development expenditure to its net sales. The empirical results support the technology-related peso problem hypothesis. In samples subject to such a peso problem, the returns are biased up and the variance is biased down.

Singing from the same hymn sheet? A closer look at the framing of human trafficing by counter-trafficing actors in Lao PDR

As disparities in wealth levels between and within countries become greater many poor people migrate in search of better earning opportunities. Some of this migration is legal but, in many cases, the difficulties involved in securing the necessary documentation mean that would-be migrants resort to illegal methods. This, in turn, makes them vulnerable to human trafficking, a phenomenon that has received growing attention from NGOs, governments and the media in recent years. Despite the attention being given to human trafficking, however, there remains a certain amount of confusion over what exactly it entails though it is generally understood to refer to the transportation and subsequent exploitation of vulnerable people through means of force or deception. The increased attention that has been given to the issue of human trafficking over the last decade has resulted in new discourses emerging which attempt to explain what human trafficking entails, what the root causes of the phenomenon are and how best to tackle the problem. While a certain degree of conceptual clarity has been attained since human trafficking rose to prominence in the 1990s, it could be argued that human trafficking remains a poorly defined concept and that there is frequently confusion concerning the difference between it and related concepts such as people smuggling, migration and prostitution. The thesis examines the ways in which human trafficking has been conceptualised or framed in a specific national context- that of Lao PDR. Attention is given to the task of locating the major frames within which the issue has been situated, as well as considering the diagnoses and prognoses that the various approaches to trafficking suggest. The research considers which particular strands of trafficking discourse have become dominant in Lao PDR and the effect this has had on the kinds of trafficking interventions that have been undertaken in the country. The research is mainly qualitative and consists of an analysis of key texts found in the Lao trafficking discourse.