Thesis (Ph. D.)--University of Texas at Austin, 2006. / Vita. Includes bibliographical references.

No description available.

Doctor of Philosophy / Psychological Sciences / Patrick A. Knight / In this research, two studies were conducted to examine the factors influencing reliance on a decision aid in personnel selection decisions. Specifically, this study examined the effect of feedback, the validity of selection predictors, and the presence of a decision aid on the use of the decision aid in personnel selection decisions. The results of both studies demonstrate that when people are provided with the decision aid, their predictions were significantly more similar to (but not the same as) the predictions made by the aid than people who were not provided with the decision aid. This suggests that when people are provided with an aid, they will use it at least to some degree. This research also shows that when provided with a decision aid that has high validity, people will increase their reliance on the decision aid over multiple decisions. Finally, this research shows that, in general, there are individual differences that influence how participants weight the different selection predictors.

The Bayesian theory has long been the predominate method in dealing with uncertainties in civil engineering practice including water resources engineering. However, it imposes unnecessary restrictive requirements on inferential problems. Concerns thus arise about the effectiveness of using Bayesian theory in dealing with more general inferential problems. The recently developed Dempster-Shafer theory appears to be able to surmount the limitations of Bayesian theory. The new theory was originally proposed as a pure mathematical theory. A reasonable amount of work has been done in trying to adopt this new theory in practice, most of this work being related to inexact inference in expert systems and all of the work still remaining in the fundamental stage. The purpose of this research is first to compare the two theories and second to try to apply Dempster-Shafer theory in solving real problems in water resources engineering. In comparing Bayesian and Dempster-Shafer theory, the equivalent situation between these two theories under a special situation is discussed first. The divergence of results from Dempster-Shafer and Bayesian approaches under more general situations where Bayesian theory is unsatisfactory is then examined. Following this, the conceptual difference between the two theories is argued. Also discussed in the first part of this research is the issue of dealing with evidence including classifying sources of evidence and expressing them through belief functions. In attempting to adopt Dempster-Shafer theory in engineering practice, the Dempster-Shafer decision theory, i.e. the application of Dempster-Shafer theory within the framework of conventional decision theory, is introduced. The application of this new decision theory is demonstrated through a water resources engineering design example. / Applied Science, Faculty of / Civil Engineering, Department of / Graduate

Decision analysis is a quantitative approach to decision-making that could bridge the gap between decisions based solely on evidence and the unique values and preferences of individual patients, a feature especially important when existing clinical evidence cannot support clear recommendations and there is a close balance between harms and benefits for the treatment options under consideration. Low molecular weight heparin for the prevention of venous thromboembolism (VTE) during pregnancy represents one such situation. The objective of this thesis is to explore the use of a decision analysis intervention for shared decision-making for thromboprophylaxis during pregnancy. This thesis begins with a scoping review that explores the ways in which decision analysis has been used to inform shared decision-making encounters, highlighting key challenges for implementing and evaluating this type of intervention. This is followed by a protocol that presents the methodology of an explanatory sequential mixed methods pilot study for the Decision Analysis in SHared decision making for Thromboprophylaxis during Pregnancy (DASH-TOP) tool. This tool was pilot tested through interviews of eligible women in Canada and Spain who were facing the treatment decision for the prevention of VTE in the antenatal period. While the tool was well received by patients, more effective ways of obtaining patient preferences and presenting the decision analysis results are required to enhance shared decision-making interactions. Finally, this thesis concludes with a reflection on the lessons learned from developing and evaluating a decision analysis intervention for shared decision-making. The insights from this research have informed the development of an integrated online shared decision-making tool for VTE in the antenatal period, which the DASH-TOP team plans to evaluate in a randomized controlled trial. It is hoped that this information will also provide guidance to researchers interested in developing or evaluating decision analysis interventions for other clinical decisions. / Dissertation / Doctor of Philosophy (PhD)

Typescript (photocopy). / Digitized by Kansas Correctional Industries

This research work focuses on the optimal algorithms of decision making and forecasting respectively, in order to achieve a better prediction. Decision making techniques and forecasting methods are investigated due to the poor accuracy of forecasting in comparison with real world data. The uncertainty of real world data leads to the use of type-1 fuzzy sets, type-2 fuzzy sets, fuzzy decision tree and fuzzy time-series for fuzzy data-mining - to which they are applied for the look-ahead based interval-valued fuzzy decision tree with optimal perimeter of the neighbourhood (LAIVFDT-OPN) model, and high-order type-2 fuzzy time series (HO-T2FTS) model. In the experiment with a real world business, a ‘computerised maintenance integration management system’ (CMIMS) is constructed as a simulation model for a case study. The CMIMS model consists of the LAIVFDT-OPN and HO-T2FTS models. It is also applied to a set of real world data from a factory in Thailand. Due to the significant uncertainty involved in machine maintenance, most tasks in machine diagnosis are still carried out manually by technicians. In this research, a prototype of CMIMS employing fuzzy data mining to diagnose machine maintenance is constructed. Considering the special features of machine maintenance data, fuzzy decision trees and fuzzy time series are adopted in the proposal method. To represent the uncertain fuzzy memberships, interval-valued fuzzy decision trees are proposed and an optimal neighbourhood perimeter is defined for look-ahead fuzzy decision trees. Based on the existing first-order type-2 time-series and high-order type-1 fuzzy time series, an improved high-order type-2 fuzzy time series method is put forward. In this case study, the CMIMS model can be used to analyse and evaluate uncertain data. It also can be employed to facilitate decision making in machine equipment status, and forecast machine maintenance plan in the future in stead of technicians. Our results demonstrated that the proposal method is effective in fuzzy decision support for machine maintenance.

Decision theory suggests two general ways to study decisions: t e process y lich the decision was made or the attributes of the decision itself. Within the process branch, Allison (1971) and Allen (1979) analyzed a single ~cision using the Rational model, the Organizational Process model, and the Jreaucratic Politics model, each of which contributed to explaining the decision. Llison and Allen also argued for how their findings could be applied more enerally. Aharoni (1966) and Heclo and Wildavsky (1974) used a second approach t :udy pieces of many decisions. The emphasis was on understanding the whole proces ather than on explaining any particular decision. This approach is largely free f decision models and describes important factors within the process. One set of attributes comes from Cooke and Slack (1984) who suggested that !cisions can be analyzed by three related pairs of variables: Operational - :rategic, Independent - Dependent, and Structured - Unstructured. To these, the esearcher added Relative Importance and Innovation. This paper uses both the process and attribute approaches to analyze eleven cas ~udies of local capital allocation decisions in two health districts in northern ngland. Some of the cases were chosen to allow direct comparison between rganizations, circumstances, and individuals. Like Allison or Allen, each decisio as examined through each model (Hypothesis 1: Each decision model helps explain ach decision). Following Aharoni or Heclo and Wildavsky. the examination of many ecisions allows delineation of important factors within the process (Hypothesis 2 here are specific factors which are important to explain the process). Finally, ach decision is also classified using Cooke and Slack's decision attributes Hypothesis 3: Decisions can be explained using the various decision attributes). Information was gathered through interviews with participants and examination 0 elevant documents. Some of the decisions were concurrent with the researcher's tay and could be watched as they happened. However, most events occurred before his involvement. Part way into the study, it became clear that each district had an rganizational process for the kinds of decisions in the study. This process was ased on three elements: the rules and regulations within the NHS which prescribe hat is acceptable and how it is to be done; past events and personalities within he health district itself which influenced current processes; and events outside he NHS which established the climate within which all of this took place. etailing this process became the basis for Part II. Because an organizational process describes the overall approach, the Organizaional Process model should also best explain the case study decisions (Hypothesis ). This alternative to Hypothesis 1 was tested in Part III, where each case stud S described and analyzed. However, the results show that the Bureaucratic Politic ~del is also important in explaining the decisions, with the Rational model less ,aeful. Thus, neither Hypothesis 1 nor Hypothesis 4 is totally supported. As expected in Hypothesis 2, many factors affected decisions (see Part IV), som pecific to the setting and some with broader applicability. The more interesting nes are Staff and Patient Welfare, the Availability of Money, and the Threshhold :ffect. Also in Part IV, the results from Part III are combined in a matrix, with eciaion models on one axis and decision attributes on the other, to analyze how :he two approaches are related. The results show some of the expected relationship letween models and attributes on a broad basis but not on a detailed basis. Significance: This study broadens decision theory by taking a first step in :ombining the attribute and process approaches. It also illuminates the need for :urther study by showing how incompatible the results of the twosaPDrQaches are. ee "Notes on reverse

This thesis examines the impact of taxation and financial factors on the level of investment in fixed assets by quoted manufacturing companies in the United Kingdom between 1971 and 1986. Its most important theme is that there exist substantial differences between companies in the way that they are affected by both taxation and financial factors. The empirical work therefore uses individual company accounting and stock market data (described in Appendix A) together with a detailed model of the corporation tax system (described in Appendix B) in order to exploit cross sectional as well as time series variation. Chapters 2 and 3 investigate the role played by taxation in the investment decision. Part of the cross sectional variation in taxation arises through 'tax exhaustion', caused by the asymmetric treatment of taxable profit and loss in UK corporation tax and restrictions on the use of the imputation system. Two investment equations, the first based on Tobin's Q and the second on the cost of capital in an Euler equation framework are developed from the same neoclassical model of the firm which explicitly models tax exhaustion and the role played by expectations. Each is a forward-looking model, which could be used for the purposes of simulating the effects of tax reform on investment, whether the reform is announced or unannounced, permanent or temporary. The results confirm that tax does play a role in the determination of investment, although, for various reasons, the precise effect is difficult to quantify. They also suggest that the Q model is a poor means of assessing the impact of taxation on investment and that it is dominated by the second model. Chapters 2 and 3 also consider the impact of taxation on company financial policy, and, in particular consider various regimes in which the company may find itself which depend on tax exhaustion and agency costs of debt. The stability of these regimes is more complex than commonly argued in the literature. The appropriate definition of the cost of capital is also developed further, under similar conditions, and a matrix of nine possible values is constructed, depending on the marginal source of finance in this period and the next period. Chapter 4 discusses the role played by financial factors. A model with legal constraints on financial behaviour and agency costs on debt is developed which predicts that, for all firms, investment depends on the level of cash generated, as well as Tobin's Q. The importance of cash flow for firms of different size and age is investigated. The results support the hypothesis that cash flow is a significant determinant of investment for all firms. Cash flow has the highest impact for large and new firms.

No description available.