Browsing by Subject "predictors"
Now showing 1 - 2 of 2
Results Per Page
Sort Options
Item An analysis of selected pre- and post-admission variables as they relate to the retention of new freshmen at a large, research, public university(Texas A&M University, 2004-09-30) Boyd, Kriss HopeTexas A&M University changed the criteria for freshman admission after a legal decision in 1996 removed ethnicity from the list of possible admission criteria. The process now includes subjective criteria such as activities, leadership, service and awards as well as the traditional objective criteria such as test scores and rank in high school class. The purpose of this study was to analyze the relationship between some of the admission criteria and retention of freshmen from the first fall to the second fall. Retention of freshmen is a performance indicator for higher education in Texas. The results of the logistic regressions showed that the relationships were modest at best and had a very small pseudo r2. The objective criteria of test scores and high school rank were either not significant or did almost nothing to increase the odds ratio. The only variable that was significant in the regression, but had a modest odds ratio, across the regression for all students and for the regressions for the subgroups of female and male students, Anglo, Hispanic and Asian American students, and for students from targeted, disadvantaged high schools was parents' education level. The points assigned to students by admissions counselors for self-reported leadership activities were significant for the regressions for all students, for female students and for Anglo students, but did very little to increase the likelihood of retention. Test scores were significant in the regressions for all students, for female students and for Hispanic students, but did almost nothing to increase the likelihood of retention. None of the variables were significant in the regression for the small group of African American students. One conclusion from the analysis is that some students whose parents have the lowest levels of education and some students from targeted high schools have unmet needs that cause higher attrition rates for these groups. However, even within these groups, there are other factors driving the students' commitment to stay enrolled for the second year at the institution than those included in this study.Item Empirical modeling of end-to-end delay dynamics in best-effort networks(Texas A&M University, 2005-08-29) Doddi, SrikarQuality of Service (QoS) is the ability to guarantee that data sent across a network will be recieved by the desination within some constraints. For many advanced applications, such as real-time multimedia QoS is determined by four parameters--end-to-end delay, delay jitter, available bandwidth or throughput, and packet drop or loss rate. It is interesting to study and be able to predict the behavior of end-to-end packet delays in a Wide area network (WAN) because it directly a??ects the QoS of real-time distributed applications. In the current work a time-series representation of end-to-end packet delay dynamics transported over standard IP networks has been considered. As it is of interest to model the open loop delay dynamics of an IP WAN, the UDP is used for transport purposes. This research aims at developing models for single-step-ahead and multi-step-ahead prediction of moving average, one-way end-to-end delays in standard IP WAN??s. The data used in this research has been obtained from simulations performed using the widely used simulator ns-2. Simulation conditions have been tuned to enable some matching of the end-to-end delay profiles with real traffic data. This has been accomplished through the use of delay autocorrelation profiles. The linear system identification models Auto-Regressive eXogenous (AR) and Auto-Regressive Moving Average with eXtra / eXternal (ARMA) and non-linear models like the Feedforwad Multi-layer Perceptron (FMLP) have been found to perform accurate single-step-ahead predictions under varying conditions of cross-traffic flow and source send rates. However as expected, as the multi-step-ahead prediction horizon is increased, the models do not perform as accurately as the single-step-ahead prediction models. Acceptable multi-step-ahead predictions for up to 500 msec horizon have been obtained.