Matching Items (3)
Filtering by
- All Subjects: Monitoring
- Creators: Bates, Thomas
- Creators: Bekki, Jennifer
- Resource Type: Text
Description
Technological advances have enabled the generation and collection of various data from complex systems, thus, creating ample opportunity to integrate knowledge in many decision making applications. This dissertation introduces holistic learning as the integration of a comprehensive set of relationships that are used towards the learning objective. The holistic view of the problem allows for richer learning from data and, thereby, improves decision making.
The first topic of this dissertation is the prediction of several target attributes using a common set of predictor attributes. In a holistic learning approach, the relationships between target attributes are embedded into the learning algorithm created in this dissertation. Specifically, a novel tree based ensemble that leverages the relationships between target attributes towards constructing a diverse, yet strong, model is proposed. The method is justified through its connection to existing methods and experimental evaluations on synthetic and real data.
The second topic pertains to monitoring complex systems that are modeled as networks. Such systems present a rich set of attributes and relationships for which holistic learning is important. In social networks, for example, in addition to friendship ties, various attributes concerning the users' gender, age, topic of messages, time of messages, etc. are collected. A restricted form of monitoring fails to take the relationships of multiple attributes into account, whereas the holistic view embeds such relationships in the monitoring methods. The focus is on the difficult task to detect a change that might only impact a small subset of the network and only occur in a sub-region of the high-dimensional space of the network attributes. One contribution is a monitoring algorithm based on a network statistical model. Another contribution is a transactional model that transforms the task into an expedient structure for machine learning, along with a generalizable algorithm to monitor the attributed network. A learning step in this algorithm adapts to changes that may only be local to sub-regions (with a broader potential for other learning tasks). Diagnostic tools to interpret the change are provided. This robust, generalizable, holistic monitoring method is elaborated on synthetic and real networks.
The first topic of this dissertation is the prediction of several target attributes using a common set of predictor attributes. In a holistic learning approach, the relationships between target attributes are embedded into the learning algorithm created in this dissertation. Specifically, a novel tree based ensemble that leverages the relationships between target attributes towards constructing a diverse, yet strong, model is proposed. The method is justified through its connection to existing methods and experimental evaluations on synthetic and real data.
The second topic pertains to monitoring complex systems that are modeled as networks. Such systems present a rich set of attributes and relationships for which holistic learning is important. In social networks, for example, in addition to friendship ties, various attributes concerning the users' gender, age, topic of messages, time of messages, etc. are collected. A restricted form of monitoring fails to take the relationships of multiple attributes into account, whereas the holistic view embeds such relationships in the monitoring methods. The focus is on the difficult task to detect a change that might only impact a small subset of the network and only occur in a sub-region of the high-dimensional space of the network attributes. One contribution is a monitoring algorithm based on a network statistical model. Another contribution is a transactional model that transforms the task into an expedient structure for machine learning, along with a generalizable algorithm to monitor the attributed network. A learning step in this algorithm adapts to changes that may only be local to sub-regions (with a broader potential for other learning tasks). Diagnostic tools to interpret the change are provided. This robust, generalizable, holistic monitoring method is elaborated on synthetic and real networks.
ContributorsAzarnoush, Bahareh (Author) / Runger, George C. (Thesis advisor) / Bekki, Jennifer (Thesis advisor) / Pan, Rong (Committee member) / Saghafian, Soroush (Committee member) / Arizona State University (Publisher)
Created2014
Description
This paper examines how equity analysts' roles as information intermediaries and monitors affect corporate liquidity policy and its associated value of cash, providing new evidence that analysts have a direct impact on corporate liquidity policy. Greater analyst coverage (1) reduces information asymmetry between a firm and outside shareholders and (2) enhances the monitoring process. Consistent with these arguments, analyst coverage increases the value of cash, thereby allowing firms to hold more cash. The cash-to-assets ratio increases by 5.2 percentage points when moving from the bottom analyst-coverage decile to the top decile. The marginal value of $1 of corporate cash holdings is $0.93 for the bottom analyst-coverage decile and $1.83 for the top decile. The positive effects remain robust after a battery of endogeneity checks. I also perform tests employing a unique dataset that consists of public and private firms, as well as a dataset that consists of public firms that have gone private. A public firm with analyst coverage can hold approximately 8% more cash than its private counterpart. These findings constitute new evidence on the real effect of analyst coverage.
ContributorsChang, Ching-Hung (Author) / Bates, Thomas (Thesis advisor) / Bharath, Sreedhar (Committee member) / Lindsey, Laura (Committee member) / Arizona State University (Publisher)
Created2012
Description
By matching a CEO's place of residence in his or her formative years with U.S. Census survey data, I obtain an estimate of the CEO's family wealth and study the link between the CEO's endowed social status and firm performance. I find that, on average, CEOs born into poor families outperform those born into wealthy families, as measured by a variety of proxies for firm performance. There is no evidence of higher risk-taking by the CEOs from low social status backgrounds. Further, CEOs from less privileged families perform better in firms with high R&D spending but they underperform CEOs from wealthy families when firms operate in a more uncertain environment. Taken together, my results show that endowed family wealth of a CEO is useful in identifying his or her managerial ability.
ContributorsDu, Fangfang (Author) / Babenko, Ilona (Thesis advisor) / Bates, Thomas (Thesis advisor) / Tserlukevich, Yuri (Committee member) / Wang, Jessie (Committee member) / Arizona State University (Publisher)
Created2018