In order to reduce default risk, financial institutions have been investigating into credit ratings of companies, which they want to give credit to. This research tries to give a method for financial institutions to differentiate between default and normal company with financial ratios, which is already announced in their seasonal financial reports. The samples are abstracted from security markets, and restricted to building companies. With Discriminant analysis and Logistic regression models, financial institutions can estimate what company may become into default situation and others stay in good condition. According to this research, financial ratios that can be used to discriminate between default and normal companies are: net worth ratio and short-turn borrowing/liquid asset and asset turnover and gross profit margin. It can also be described with asset turnover and gross profit margin if default risk is been estimated.

In recent years, artificial neural networks (ANNs) have been employed for estimation and prediction purposes in many areas of civil/structural engineering. In this thesis, multilayered feedforward backpropagation algorithm is used for the approximate analysis and calibration of RC T-beam bridges and modeling of bridge ratings of these bridges. Currently bridges are analyzed using a standard FEM program. However, when a large population of bridges is concerned, such as the one considered in this project (Pennsylvania T-beam bridge population), it is impractical to carry out FEM analysis of all bridges in the population due to the fact that development and analysis of every single bridge requires considerable time as well as effort. Rapid and acceptably approximate analysis of bridges seems to be possible using ANN approach. First part of the study describes the application of neural network (NN) systems in developing the relationships between bridge parameters and bridge responses. The NN models are trained using some training data that are obtainedfrom finite-element analyses and that contain bridge parameters as inputs and critical responses as outputs. In the second part, ANN systems are used for the calibration of the finite element model of a typical RC T-beam bridge -the Manoa Road Bridge from the Pennsylvania&rsquo / s T-beam bridge population - based on field test data. Manual calibration of these models are extremely time consuming and laborious. Therefore, a neural network- based method is developed for easy and practical calibration of these models. The ANN model is trained using some training data that are obtained from finite-element analyses and that contain modal and displacement parameters as inputs and structural parameters as outputs. After the training is completed, fieldmeasured data set is fed into the trained ANN model. Then, FE model is updated with the predicted structural parameters from the ANN model. In the final part, Neural Networks (NNs) are used to model the bridge ratings of RC T-beam bridges based on bridge parameters. Bridge load ratings are calculated more accurately by taking into account the actual geometry and detailing of the T-beam bridges. Then, ANN solution is developed to easily compute bridge load ratings.

Taiwanese enterprises in Taiwan, China, and Hong Kong operate under division of labor through perpendicular and level integration and strategic alliances. As Mainland China has becoming a mature economic entity, the strategy has changed from the previous ¡§Soliciting Business and Introducing Investment¡¨ to ¡§Soliciting Business and Selecting Investment¡¨. In 2007, China announced the foreign enterprises investment and industry directives, and revoked tax incentives. In 2008, due to the levy of corporate income tax and the execution of the Labor Contract Law, the operating cost for businesses has increased, and the competitiveness of Chinese labors has diminished. Thus, Taiwanese enterprises have to find cheap resources elsewhere, and Vietnam is then recognized as a popular major investment location by Taiwanese enterprises. According to the Financial Supervisory Commission, Executive Yuan, the pre-tax profit for the Taiwan domestic banks in 2007 was NT $38.9 billion. Among which, overseas bank branches (including Offshore Banking Unit) had NT $31.1 billion, accounting for 79.9% of the total profit. It shows that overseas bank branches are a major source of profit for domestic banks, and the profit is growing rapidly year after year. Hence, domestic banks have actively established overseas offices and branches for business development, and followed the steps of Taiwanese enterprises to provide loans services and create higher profits for the banks. This study adopted a qualitative research method and designed an in-depth interview with 30 enterprise owners and senior financial supervisors. Questionnaire survey was conducted to analyze bank loans strategy, competitive strategy SWOT analysis, and AHP analysis. Research results showed that the most important factor of an ideal Taiwan-based bank to Taiwanese enterprises is the interest of loan. Besides being a reliable long-term partner with Taiwanese enterprises, banks should pay more attention to their enterprise credit ratings and strengthen risk management. To help banks achieving the win-win situation by optimizing the operating performance and maximizing the profits for banks, this study proposed several suggestions concerning loans strategy for Taiwan-based bank.

This paper will use political discussion programs in Taiwan as an example and use the research method of experimentation to attempt to establish a way of research that is both quantitative and qualitative. By using a correlation analysis of audience evaluation and ratings, an investigation will be carried out as to whether Taiwan¡¦s political discussion programs are in fact public forums that are a paradigm of a deliberative democracy. In order to test the hypothesis that states there is a correlation between audience evaluation and ratings, this paper made audience evaluation (in other words the topics discussed on the programs, their hosts, guests, producers and their negative effects) an independent variable and made the AGB Nielsen television rating numbers a dependent variable. It was discovered that audience evaluation does in fact affect ratings and that there is a positive correlation between audience evaluation and ratings. For example, the higher an audience¡¦s evaluation of the topics discussed on a certain program and its hosts, guests and producers were, the higher that particular program¡¦s ratings were. If on the other hand, audiences believe that a certain political discussion program has negative effects such as encouraging opposition within society or defaming a certain person or group, the lower the ratings of that particular program were. From the results of the correlation analysis of audience evaluation and ratings, it can be seen that there is a huge difference between what viewers expect from political discussion programs and what our current political discussion programs are like. It was also discovered that whilst today¡¦s political discussion programs claim they are public forums; they are in reality primarily aimed at making commercial profit. This means that these programs are still a long way away from becoming a paradigm for a deliberative democracy and making Taiwan a more democratic society.

Previous research suggests that discrepant ratings of symptoms, behaviors, and competencies may have important implications for childhood adjustment. Consequently, several theoretical explanations regarding the meanings and implications of inter-rater discrepancies have been proposed. The current study examined several largely unexplored issues regarding the statistical and conceptual properties of discrepant ratings; these include heterogeneity of agreement, shared method variance, and direction-specific effects. In a sample of 384 seventh and eighth-grade adolescents, it was found that each of the issues is integral in the interpretation of significant relationships between discrepant ratings of social rejection and childhood adjustment variables. More specifically, results demonstrated that the influence of shared method variance appears to be largely related to heterogeneity of agreement, and that accounting for shared method variance can increase, decrease, or even change the direction of the relationship between discrepancy scores and other variables. Also, results showed that the magnitude of discrepancies can be unrelated, equally related, or differentially related to the adjustment variables for the different forms of bias. Altogether, the findings of this study have implications for the theoretical conceptualization and statistical analysis of inter-rater discrepancy scores.

In this dissertation, I explore the ability of financial institutions to impact firm behavior. The first essay examines whether relationships between venture capital owners (VCs) and investment banking underwriters affect a firm’s ability to issue equity. I find that past interactions between VC owners and underwriters in the form of previously underwritten initial public offerings (IPOs) significantly increase the likelihood that an IPO firm chooses a specific underwriter. In terms of how VCs and underwriters associate with each other, older VCs partner with more reputable underwriters. Despite paying higher fees, issuing firms benefit from stronger VC-Underwriter relationships through upward offer price revisions and higher valuations. VC-Underwriter relationships also predict underwriter choice in subsequent equity offerings. This essay provides empirical evidence that suggests VCs use their relationships with investment banks to enable their portfolio firms access to high quality underwriters and better underwriting services. The second essay investigates whether credit rating concerns affect capital investment decisions. Using ex-ante measures of proximity to a rating change, I find that firms that are near a credit rating downgrade spend significantly less on capital expenditures relative to those not near a rating change. The response of firms to credit rating upgrades is not symmetric: firms do not seem to adopt significantly different investment policies when near an upgrade. This effect of lowering investment when near a credit downgrade is stronger for firms that face financial constraints, experience greater adverse selection and are more active in debt markets. Related to reductions in investment, firms near rating changes also spend less on research and development expenses and pay lower dividends. My findings are consistent with firms conserving financial resources to prevent adverse credit rating changes that could increase their cost of capital. / text

This study examined whether hiring recommendations based on thin slices of a structured interview were associated with recommendations based on the interview in its entirety. After viewing 12 seconds of silent interviewee behavior, participants made hiring recommendations that were significantly correlated with those produced by individuals viewing a still-frame of the interview and the entire interview. In an effort to determine what sources of information participants were using to arrive at their recommendations, nonverbal behaviors were examined in detail. Applicants who appeared attentive, not anxious, competent, confident, dominant, optimistic, and professional were more likely to receive positive hiring recommendations than others. Additional analyses reveal that these nonverbal behaviors impact hiring recommendations in both the still-frame and thin-slice video conditions after controlling for applicant physical attractiveness. Overall, results indicate that despite the availability of verbal content, interviewers may be heavily influenced by their first 12-second impression of a job applicant.

This study built upon the theoretical feedback process of Ilgen, Fisher and Taylor (1979), as refined by Kinicki, Prussia, Wu, & Mckee-Ryan (2004), to contribute to a more comprehensive understanding of the mechanism underlying an individual's response to performance feedback. The feedback model implicates source credibility and the feedback-rich environment as critical elements in the process explaining recipient accuracy perceptions and responses. Thus the sign and perceived accuracy of performance feedback were investigated in a 3 x 3 (plus control) experimental design. One hundred fifty-six undergraduate students were randomly assigned to feedback conditions, in which they performed on a three-trial Lego model reproduction task. After trial one, participants received false feedback from a confederate supervisor. The feedback was based on fictional norm tables, which framed the participant's performance as falling into the 70th percentile (positive), 50th percenti le (average), or 30th percentile (negative) according to condition. The supervisor then supplemented the norm tables with conclusive comments designed to be perceived as positive distortion of the norm table feedback, reinforcement of norm tables (accurate feedback), or negative distortion of the norm tables. Performance time, errors, task-specific self-efficacy, self-assessment of performance, and self-report effort-expended data were collected on trials one through three. The results indicated a significant interaction between feedback sign and perceived feedback accuracy on participant performance (F(4,132) = 3.72, p < 0.01), whereby the performance in the positively distorted-positive sign feedback condition was significantly higher than performance in the positively distorted-negative sign feedback condition. When the feedback was perceived to be accurate, negative sign feedback resulted in significantly better performance compared to the positive feedback condition (contrar y to previous research). Task-specific self-efficacy was not found to mediate the relationship between type of feedback and performance, and no significant effect of feedback sign or perceived feedback accuracy on task-specific self-efficacy was found. These findings provide possible explanation as to why supervisors tend to positively distort both sign and accuracy in performance appraisals (e.g., Benedict & Levine, 1988). Implications for theoretical expansion of the feedback process model, and application to workplace performance management are discussed.

Following the early days of the Internet and the World Wide Web, news media in Canada have gone on to develop their own news web sites with the intentions of meeting the on-line needs of media audiences, expanding their audience reach, and adding to revenue production and profitability on- and off-line. Web strategies have varied somewhat across the different media, but anecdotal evidence suggests that sports contents have been important for both print and television. This thesis focused on the latter, sports contents on television network websites, and was undertaken to evaluate how Canadian television stations are utilizing the Internet and web technologies to feature sports news and information. Only a few studies specific to sports television web sites have been done, and these have mainly focused on American news stations. The research objective of the thesis was to systematically examine the web presence of sports contents on Canadian television web sites by conducting a content analysis of identifiably unique sites in the Canadian context. A site analysis protocol was developed through an iterative process. An initial instrument was constructed drawing on past research in this area. In particular, prior work by Bates et al. (1996 & 1997), Pines (1999), Bucy, Lang, Potter & Grabe (1999), Sparks (2001) provided systematic measures for examiriirig the Web presence of television stations. Ha & James's definition of interactivity (1998) was also useful as was the work of Cho (1999), Rogers & Thorson (2000) on Internet advertising. The initial instrument was evaluated and modified during a series of trial scans. The final instrument focused on five areas: body of the home page, types of content, presentation mechanisms, interactivity and advertising. A systematic site analysis was conducted from August to October, 2003, and a total of twenty-one sports home pages were analyzed. Three web sites (TSN, Leafs TV and The Score) were found to have a good balance in the five areas evaluated in the study. The results of independent-samples t-tests showed that general television networks had more sports top news and hyperlinks to other news items than sport specialty networks. By comparison, sports specialty networks tended to have more sport-related search engines and greater efficiency of space. CBC's "Sports Forums" that were configured on its sports home page gave the public broadcaster the highest quotient for interactivity in comparison with the twenty private networks and stations in the study. Advertising was present in all of the sites, and the findings point to an increasing interest in the televisual and sport web site media in producing revenue through web-based advertising.

This paper introduces a model framework for dynamic credit rating processes. Our framework aggregates ordinal rating information stemming from a variety of rating sources. The dynamic of the consensus rating captures systematic as well as idiosyncratic changes. In addition, our framework allows to validate the different rating sources by analyzing the mean/variance structure of the rating errors. In an empirical study for the iTraxx Europe companies rated by the big three external rating agencies we use Bayesian techniques to estimate the consensus ratings for these companies. The advantages are illustrated by comparing our dynamic rating model to a benchmark model. (author´s abstract) / Series: Research Report Series / Department of Statistics and Mathematics