No description available.

No description available.

Cure rate models are widely used to model time-to-event data in the presence of long-term survivors. Cure rate models, since introduced by Boag (1949), have gained significance over time due to remarkable advancements in the drug industry resulting in cures for a number of diseases. In this thesis, cure rate models are considered under a competing risk scenario wherein the initial number of competing causes is described by a Conway-Maxwell (COM) Poisson distribution, under the assumption of proportional hazards (PH) lifetime for the susceptibles. This provides a natural extension of the work of Balakrishnan & Pal (2013) who had considered independently and identically distributed (i.i.d.) lifetimes in this setup. By linking covariates to the lifetime through PH assumption, we obtain a flexible cure rate model. First, the baseline hazard is assumed to be of the Weibull form. Parameter estimation is carried out using EM algorithm and the standard errors are estimated using Louis' method. The performance of estimation is assessed through a simulation study. A model discrimination study is performed using Likelihood-based and Information-based criteria since the COM-Poisson model includes geometric, Poisson and Bernoulli as special cases. The details are covered in Chapter 2. As a natural extension of this work, we next approximate the baseline hazard with a piecewise linear functions (PLA) and estimated it non-parametrically for the COM-Poisson cure rate model under PH setup. The corresponding simulation study and model discrimination results are presented in Chapter 3. Lastly, we consider a destructive cure rate model, introduced by Rodrigues et. al (2011), and study it under the PH assumption for the lifetimes of susceptibles. In this, the initial number of competing causes are modeled by a weighted Poisson distribution. We then focus mainly on three special cases, viz., destructive exponentially weighted Poisson, destructive length-biased Poisson and destructive negative binomial cure rate models, and all corresponding results are presented in Chapter 4. / Thesis / Doctor of Philosophy (PhD)

Scheduling forms an important aspect of operating systems because it has a direct impact on system performance. Most existing general-purpose schedulers use a priority-based scheme to schedule processes. Such priority-based mechanisms cannot guarantee proportional fairness for every process. Proportional share schedulers maintain fairness among tasks based on given weight values. In both of these scheduler types, the scheduling decision is done per-process. However, system usage policies are typically set on a per-consumer basis, where a consumer represents a group of related processes that may belong to the same application or user. The COPS framework uses the idea of consumer sets to group processes. Its design guarantees system usage per consumer, based on relative weights. We have added a share management layer on top of a proportional share scheduler to ease the administrative job of share assignment for these consumer sets. We have evaluated our system in real world scenarios and show that the CPU usage for consumer sets with CPU-bound processes complies with the administrator-defined policy goals. / Master of Science

The Vaccine Adverse Event Reporting System (VAERS) received thousands of reports of adverse events that occurred after vaccine administrations from the post-marketing vaccine safety surveillance. However, the causality between vaccines and reported adverse events cannot be taken for granted. In this report several data mining methods were applied to VAERS database that is coded in MedDRA terms to discover possible associations between vaccines and adverse events. Efforts were devoted to identify events that are reported more frequently after administering one vaccine than other vaccines using the following data mining techniques: relative ratio (RR), statistical significance (LogP), proportional reporting ratio (PRR), and screened PRR (SPRR). The vaccine-event combinations that ranked top in each method varied substantially among the methods. RR and PRR gave excessive weight to small counts of vaccine-event pairs, but SPRR was able to correct this weakness. There are only 33 vaccine-event pairs that were shared among the top 1,000 ranked in each method. Evaluating the properties of these data mining methods and exploring other methods will help improve vaccine safety surveillance. / text

In 2005 Prince Edward Island’s plebiscite on replacing the single member plurality (SMP) voting system with a mixed member proportional (MMP) voting system was defeated. In New Brunswick a similar referendum, recommended by the Commission on Legislative Democracy in 2004, was never held. This thesis investigates media coverage of these recent electoral reform initiatives in Prince Edward Island and New Brunswick. Specifically, it examines local newspaper coverage of each province’s electoral reform initiatives and analyses the findings to determine if the newspapers demonstrated any bias for or against the implementation of a new voting system. It concludes that in each province the local newspaper media demonstrated a pro-electoral reform position, though there was considerable variation between the newspapers with respect to the breadth of coverage, the amount of bias and how that bias was articulated to readers. / Graduate

Considerable amounts of event history data are collected through longitudinal surveys. These surveys have many particularities or features that are the results of the dynamic nature of the population under study and of the fact that data collected through longitudinal surveys involve the use of complex survey designs, with clustering and stratification. These particularities include: attrition, seam-effect, censoring, left-truncation and complications in the variance estimation due to the use of complex survey designs. This thesis focuses on the last two points. Statistical methods based on the stratified Cox proportional hazards model that account for intra-cluster dependence, when the sampling design is uninformative, are proposed. This is achieved using the theory of estimating equations in conjunction with empirical process theory. Issues concerning analytic inference from survey data and the use of weighted versus unweighted procedures are also discussed. The proposed methodology is applied to data from the U. S. Survey of Income and Program Participation (SIPP) and data from the Canadian Survey of Labour and Income Dynamics (SLID). Finally, different statistical methods for handling left-truncated sojourns are explored and compared. These include the conditional partial likelihood and other methods, based on the Exponential or the Weibull distributions.

The voting system a government chooses has influence upon the way people vote, as it has some bearing on the average person’s utility received from voting. In the United States the Electoral College system is used to determine who becomes president-elect, but theoretical arguments have been made to support a change in voting structure to Proportional Representation. I developed my model to understand more about how a change in the voting structure would affect voter utility. My contribution to the research question focuses upon relative benefits of one voter in a specific party to another voter in another party. I used an empirical approach, using data pulled from past elections and survey data. My model suggests that Republican voters would receive a higher utility from a change in the voting system from the Electoral College to Proportional Representation. It is unclear whether Democrats would receive a higher benefit, but at least would receive a lesser utility change than that of the Republican voters.

Voter frustration in the US is driven largely by partisanship and gridlock in Congress. This paper seeks to understand the root causes of gridlock and look at alternative methods for eliminating it. I find that while the media focuses on polarization as the root cause of gridlock, the checks and balances system plays an equally significant role, and that the interaction between the party system and the governmental structure of the US government creates incentives that cause gridlock to form. Recent reforms have failed to successfully address gridlock because they do not change the polarized party system or the barriers to policy-making in government. After acknowledging the failure of recent reforms, I consider a new set of reforms: electoral system reform. The field of electoral systems provides many policy alternatives with profound tradeoffs, many of which make gridlock obsolete. Majoritarian systems create single-party majorities and reduce checks on majority power, allowing the majority party to implement their platform tempered through voter approval rather than checks on their power. Proportional systems retain checks and balances through the creation of governing coalitions, but the parties in power have stronger incentives to cooperate and compromise than under the current polarized US system. I propose a system for evaluating electoral systems and compare them in terms of accountability, legitimacy, effectiveness, representativeness, and complexity. Finally, I propose two electoral reforms, informed by the study of electoral systems, that are both feasible and increase the ability for third parties to gain seats in the legislature. The first, proportional representation for House members, creates small to medium-sized proportional Congressional districts at the state level to reduce the threshold for party entry. The second, Alternative Vote (ranked-choice) for Senate, proposes switching to a preference ranking system for Senate elections further remove barriers to third parties. These reforms should undermine the polarized two-party system and create new incentives for cooperation in Congress.

The Sudbury Neutrino Observatory (SNO), a heavy water Cherenkov experiment, was designed to detect solar Boron-8 neutrinos via their elastic scattering interactions on electrons, or charged current and neutral current (NC) interactions on deuterium. In the third phase of SNO, an array of Helium-3 proportional counters was deployed to detect neutrons produced in NC interactions. A simulation of the current pulses and energy spectra of the main kinds of ionization events inside these Neutral Current Detectors (NCDs) was developed. To achieve this, electron drift times in NCDs were evaluated with a Monte Carlo method, and constrained by using wire alpha activity inside the counters. The pulse calculation algorithm applies to any ionization event, and takes into account processes such as straggling, electron diffusion, and propagation through the NCD hardware. A space charge model was developed to fully explain the energy spectra of neutron and alpha events. Comparisons with data allowed the various classes of alpha backgrounds to be identified, and gave evidence for the spatial non-uniformity of Uranium-238 and Thorium-232 chain nuclei in the counter walls. The simulation was applied to determine the fractional contents of the main types of alpha backgrounds in each NCD string. The number of neutron capture events in the array was extracted via a statistical separation, using Monte Carlo generated alpha background pulse shape parameter distributions and minimal energy information. The inferred total Boron-8 solar neutrino flux is: φ_NC< = 5.74 ± 0.77 (stat) ± 0.39 (sys) x 10⁶ cm^-2s^-1 in agreement with Standard Solar predictions and previous SNO results.