Agency theories suggest that governance matters more when agency conflicts are potentially more severe. However, empirical studies often do not control for the potential severity of agency conflicts. I show that the marginal benefit of governance varies with the free cash flow level, a proxy for the potential severity of agency conflicts. As the free cash flow level increases, higher governance quality becomes incrementally more value-enhancing, and lower governance quality becomes incrementally more value-destroying. This is consistent with the hypothesis that better governance helps resolve the agency conflicts in investment decisions when a firm has more free cash flows (Jensen, 1986). This study highlights the importance of controlling for the potential severity of agency conflicts in governance studies and provides an improved method to estimate the marginal benefit of a governance mechanism.

In Chinese language, the smallest meaningful unit is a word which is composed of a sequence of characters. A Chinese sentence is composed of a sequence of words without any separation between them. In the area of information retrieval or data mining, the segmentation of a sequence of Chinese characters should be done before anyone starts to use these segments of characters. The process is called the Chinese word segmentation. The researches of Chinese word segmentation have been developed for many years. Although some recent researches have achieved very high performance, the recall of those words that are not in the dictionary only achieves sixty or seventy percent. An approach described in this paper makes use of the linear-chain conditional random fields (CRFs) to have a more accurate Chinese word segmentation. The discriminatively trained model that uses two of our proposed feature templates for deciding the boundaries between characters is used in our study. We also propose three other methods, which are the duplicate word repartition, the date representation repartition, and the segment refinement, to enhance the accuracy of the processed segments. In the experiments, we use several different approaches for testing and compare the results with those proposed by Li et al. and Lau and King based on three different Chinese word corpora. The results prove that the improved feature template which makes use of the information of prefix and postfix could increase both the recall and the precision. For example, the F-measure reaches 0.964 in the MSR dataset. By detecting repeat characters, the duplicated characters could also be better repartitioned without using extra resources. In the representation of date, the wrongly segmented date could be better repartitioned by using the proposed method which deals with numbers, date, and measure words. If a word is segmented differently from that of the corresponding standard segmentation corpus, a proper segment could be produced by repartitioning the assembled segment which is composed of the current segment and the adjacent segment. In the area of using the conditional random fields for Chinese word segmentation, we have proposed a feature template for better result and three methods which focus on other specific segmentation problems.

There have been a number of studies showing that the Conditional Reasoning Test for Aggression (CRT-A) is a valid measure of one's implicit preparedness to engage in activities that are intended to harm others. Few studies have examined the predictive power of subscales of the CRT-A. The purpose of this project is to examine the validity of the CRT-A and its subscales for predicting unnecessary corrective action requests filed in a sample of employees working in a nuclear power plant. Results indicate that the Powerlessness subscale differentiates employees who file unnecessary reports from employees who do not.

In recent years, the call for business actors to be part of collaborations addressing sustainable development has become more common. There is a consensus that no single sector alone can solve the environmental problems and poverty conditions challenging humanity. However, it is not clear if these cross-sector collaborations thrive when disasters can strike any time and when some actors are richer than others. Through a series of experiments involving threshold public goods games with stochastic shocks, this dissertation contains three related papers exploring different facets of the persistence of cooperation. The experiments were conducted in Sweden, the Philippines and South Africa, countries with varying disaster risk exposures and income structures. Cooperation in the face of disaster explores the effects of different types of uncertainties on cooperation, particularly when there is a risk for repeated disasters (i.e. losses resulting from inadequate cooperation). The results show that cooperation persists when we do not know when disasters may strike (i.e. timing), as well as when there are uncertainties on what is required to avoid the disaster (i.e. threshold) and which losses will be incurred (i.e. impact). Conditional cooperation and disaster uncertainty explores the mechanism behind the persistence of cooperation, as it investigates if conditionality continues to prevail in the face of disaster. The findings show that conditionality and free-riding attenuates while unconditional cooperation accelerates. Cooperating in an unequal and uncertain world explores what happens when inequality enters the picture. The findings reveal that cooperation remains the same when there is inequality and increases in the presence of uncertainty. The effect of uncertainty is stronger than inequality, with high unconditional cooperation and low freeriding. / <p>Diss. Stockholm :  Stockholm School of Economics, 2015</p>

This thesis investigates the Theory of Formal Discipline (TFD): the idea that studying mathematics develops general reasoning skills. This belief has been held since the time of Plato (2003/375B.C), and has been cited in recent policy reports (Smith, 2004; Walport, 2010) as an argument for why mathematics should hold a privileged place in the UK's National Curriculum. However, there is no rigorous research evidence that justifies the claim. The research presented in this thesis aims to address this shortcoming. Two questions are addressed in the investigation of the TFD: is studying advanced mathematics associated with development in reasoning skills, and if so, what might be the mechanism of this development? The primary type of reasoning measured is conditional inference validation (i.e. `if p then q; not p; therefore not q'). In two longitudinal studies it is shown that the conditional reasoning behaviour of mathematics students at AS level and undergraduate level does change over time, but that it does not become straightforwardly more normative. Instead, mathematics students reason more in line with the `defective' interpretation of the conditional, under which they assume p and reason about q. This leads to the assumption that not-p cases are irrelevant, which results in the rejection of two commonly-endorsed invalid inferences, but also in the rejection of the valid modus tollens inference. Mathematics students did not change in their reasoning behaviour on a thematic syllogisms task or a thematic version of the conditional inference task. Next, it is shown that mathematics students reason significantly less in line with a defective interpretation of the conditional when it is phrased `p only if q' compared to when it is phrased `if p then q', despite the two forms being logically equivalent. This suggests that their performance is determined by linguistic features rather than the underlying logic. The final two studies investigated the heuristic and algorithmic levels of Stanovich's (2009a) tri-process model of cognition as potential mechanisms of the change in conditional reasoning skills. It is shown that mathematicians' defective interpretation of the conditional stems in part from heuristic level processing and in part from effortful processing, and that the executive function skills of inhibition and shifting at the algorithmic level are correlated with its adoption. It is suggested that studying mathematics regularly exposes students to implicit `if then' statements where they are expected to assume p and reason about q, and that this encourages them to adopt a defective interpretation of conditionals. It is concluded that the TFD is not supported by the evidence; while mathematics does seem to develop abstract conditional reasoning skills, the result is not more normative reasoning.

Background: There are a plethora of potential statistical designs which can be used to evaluate efficacy of a novel cancer treatment in the phase II clinical trial setting. Unfortunately, there is no consensus as to which design one should prefer, nor even which definition of efficacy should be used and the primary endpoint conclusion can vary depending on which design is chosen. It would be useful if an all-encompassing methodology was possible which could evaluate all the different designs simultaneously and allow investigators an understanding of the trial results under the varying scenarios. Methods: Finite Markov chain imbedding is a method which can be used in the setting of phase II oncology clinical trials but never previously evaluated in this scenario. Simple variations to the transition matrix or end-state probability definitions can be performed which allow for evaluation of multiple designs and endpoints for a single trial. A computer program is written in R which allows for computation of p-values and conditional power, two common statistical measures used for evaluation of trial results. A simulation study is performed on data arising from an actual phase II clinical trial performed recently in which the study conclusion regarding the efficacy of the potential treatment was debatable. Results: Finite Markov chain imbedding is shown to be useful for evaluating phase II oncology clinical trial results. The R code written for evaluating the simulation study is demonstrated to be fast and useful for investigating different trial designs. Further detail regarding the clinical trial results are presented, including the potential prolongation of stable disease of the treatment, which is a potentially useful marker of efficacy for this cytostatic agent. Conclusions: This novel methodology may prove to be an useful investigative technique for the evaluation of phase II oncology clinical trial data. Future studies which have disputable conclusions might become less controversial with the aid of finite Markov chain imbedding and the possible multiple evaluations which is now viable. Better understanding of activity for a given treatment might expedite the drug development process or help distinguish active from inactive treatments

The primary aim of this thesis is the elucidation of covariate effects on the dependence structure of random variables in bivariate or multivariate models. We develop a unified approach via a conditional copula model in which the copula is parametric and its parameter varies as the covariate. We propose a nonparametric procedure based on local likelihood to estimate the functional relationship between the copula parameter and the covariate, derive the asymptotic properties of the proposed estimator and outline the construction of pointwise confidence intervals. We also contribute a novel conditional copula selection method based on cross-validated prediction errors and a generalized likelihood ratio-type test to determine if the copula parameter varies significantly. We derive the asymptotic null distribution of the formal test. Using subsets of the Matched Multiple Birth and Framingham Heart Study datasets, we demonstrate the performance of these procedures via analyses of gestational age-specific twin birth weights and the impact of change in body mass index on the dependence between two consequent pulse pressures taken from the same subject.

No description available.

This thesis deals with the use of Conditional Random Fields (CRFs; Lafferty et al. (2001)) for Natural Language Processing (NLP). CRFs are probabilistic models for sequence labelling which are particularly well suited to NLP. They have many compelling advantages over other popular models such as Hidden Markov Models and Maximum Entropy Markov Models (Rabiner, 1990; McCallum et al., 2001), and have been applied to a number of NLP tasks with considerable success (e.g., Sha and Pereira (2003) and Smith et al. (2005)). Despite their apparent success, CRFs suffer from two main failings. Firstly, they often over-fit the training sample. This is a consequence of their considerable expressive power, and can be limited by a prior over the model parameters (Sha and Pereira, 2003; Peng and McCallum, 2004). Their second failing is that the standard methods for CRF training are often very slow, sometimes requiring weeks of processing time. This efficiency problem is largely ignored in current literature, although in practise the cost of training prevents the application of CRFs to many new more complex tasks, and also prevents the use of densely connected graphs, which would allow for much richer feature sets. (For complete abstract open document)

This thesis describes the implementation of the conditional moment closure (CMC) combustion model in a numerical scheme and its application to the modelling of enclosure fires. Prediction of carbon monoxide (CO) in the upper smoke layer of enclosure fires is of primary interest because it is a common cause of death. The CO concentration cannot be easily predicted by empirical means, so a method is needed which models the chemistry of a quenched, turbulent fire plume and subsequent mixing within an enclosed space. CMC is a turbulent combustion model which has been researched for over a decade. It has provided predictions of major and minor species in jet diffusion flames. The extension to enclosure fires is a new application for which the flow is complex and temperatures are well below adiabatic conditions. Advances are made in the numerical implementation of CMC. The governing combustion equations are cast in a conserved, finite volume formulation for which boundary conditions are uniquely defined. Computational efficiency is improved through two criteria which allow the reduction in the size of the computational domain without any loss of accuracy. Modelling results are compared to experimental data for natural gas fires burning under a hood. Comparison is made in the recirculating, post-flame region of the flow where temperatures are low and reactions are quenched. Due to the spatial flux terms contained in the governing equations, CMC is able to model the situation where chemical species are produced in the high temperature fire-plume and then transported to non-reacting regions. Predictions of CO and other species are in reasonable agreement with the experimental data over a range of lean and rich hood-fire conditions. Sensitivity of results to chemistry, temperature and modelling closures is inves- tigated. Species predictions are shown to be quite different for the two detailed chemical mechanisms used. Temperature conditions within the hood effect the for- mation of species in the plume prior to quenching and subsequently species predic- tions in the post-flame region are also effected. Clipped Gaussian and ß-function probability density functions (PDFs) are used for the stochastic mixture fraction. Species predictions in the plume are sensitive to the form of the PDF but in the post-flame region, where the ß-function approaches a Gaussian form, predictions are relatively insensitive. Two models are used for the conditional scalar dissipation: a uniform model, where the conditional quantity is set equal to the unconditional scalar dissipation across all mixture fraction space; and a model which is consistent with the PDF transport equation. In the plume, predictions of minor species are sensitive to the modelling used, but in the recirculating, post-flame region species are not significantly effected.