Theses, Dissertations, and other Required Graduate Degree Essays

Receive updates for this collection

This collection contains digitized SFU theses except for those theses submitted within the last 12 months. If you cannot find the thesis you are looking for please search Recently Submitted Theses as it may be a recently submitted thesis and thus not yet available in Summit.

Incorporating statistical clustering methods into mortality models to improve forecasting performances

Author: 
Date created: 
2020-04-09
Abstract: 

Statistical clustering is a procedure of classifying a set of objects such that objects in the same class (called cluster) are more homogeneous, with respect to some features or characteristics, to each other than to those in other classes. In this project, we apply four clustering approaches to improving forecasting performances of the Lee-Carter and CBD models. First, each of four clustering methods (the Ward's hierarchical clustering, the divisive hierarchical clustering, the K-means clustering, and the Gaussian mixture model clustering) are adopted to determine, based on some characteristics of mortality rates, the number and members of age subgroups from a whole group of ages 25-84. Next, we forecast 10-year and 20-year mortality rates for each of the age subgroups using the Lee-Carter and CBD models, respectively. Finally, numerical illustrations are given with R packages "NbClust" and "mclust" for clustering. Mortality data for both genders of the US and the UK are obtained from the Human Mortality Database, and the MAPE (mean absolute percentage error) measure is adopted to evaluate forecasting performance. Comparisons of MAPE values are made with and without clustering, which demonstrate that all the proposed clustering methods can improve forecasting performances of the Lee-Carter and CBD models.

Document type: 
Graduating extended essay / Research project
File(s): 
Senior supervisor: 
Cary Chi-Liang Tsai
Department: 
Science: Department of Statistics and Actuarial Science
Thesis type: 
(Project) M.Sc.

Identification of cancer drivers, conserved alteration patterns and evolutionary trajectories in tumors

Author: 
Date created: 
2020-03-25
Abstract: 

Over the past decade, high-throughput sequencing efforts have provided an unprecedented opportunity to identify genomic alterations that can lead to changes in gene regulation, protein structure, and function. During tumor progression, cancer cells accumulate a multitude of genomic alterations, a small fraction of which provide tumor cells with selective advantage - known as "driver" alterations. However, most of them are inconsequential "passenger" alterations that are effectively neutral and greatly outnumber driver alterations. Moreover, due to a high amount of heterogeneity, alteration landscapes of tumors differ between different patients and different sites. This thesis first presents HIT'nDRIVE, a method that integrates genomic and transcriptomic data to identify a set of patient-specific, sequence-altered potential driver genes, with sufficient collective influence over dysregulated transcripts through interactome. Applied to 2200 tumors, HIT'nDRIVE revealed many potentially clinically actionable driver genes and demonstrated its robustness in selecting cancer-implicated drivers. The results also show that small network modules seeded by HIT'nDRIVE-selected drivers significantly improve classification of cancer phenotypes and drug efficacy in pan-cancer cell lines compared to alternative methods and approaches. Next, a method for detection of functionally meaningful and recurrent alteration patterns within gene interaction networks, cd-CAP, is presented. In a number of TCGA data sets, cd-CAP identified large subnetworks with identically conserved alteration patterns (across many tumor samples), that were significantly associated with patients' clinical outcome. As multi-region, time-series and single cell sequencing data become more widely available, computational methods have been developed with the goal of inferring the subclonal composition and evolutionary history of tumors from tumor biopsy sequencing data. Unfortunately, the phylogenetic trees reported for many tumor samples differ significantly from other tumors with similar characteristics. This thesis presents CONETT, the first computational method for detection of conserved trajectories of alteration events in tumor evolution. Applied to two multi-region sequencing data sets of 100 tumors each, CONETT confirms all findings of the original studies and identifies additional repeated trajectories.

Document type: 
Thesis
File(s): 
Senior supervisor: 
Martin Ester
Department: 
Applied Sciences: School of Computing Science
Thesis type: 
(Thesis) Ph.D.

Essays in instrumental variables estimators

Author: 
Date created: 
2020-04-08
Abstract: 

This PhD thesis focuses on instrumental variable models. Often, econometric models are based on orthogonality conditions used to estimate parameters of interest. The literature on such models is vast, and numerous approaches have provided consistent and asymptotically normal estimators. The three chapters presented here consider different models featuring moment conditions that are estimated. In particular, it is aimed to study the finite performances of various estimators in different contexts, in order to provide guidelines on which procedure to select according to the problem at hand. The first chapter considers Euler equations, fundamental equation in dynamic stochastic macroeconomic models. I solve a generic stochastic growth model and use its solutions to generate samples in order to study the performances of moment based estimators. The second chapter studies the widely used linear model in a context where the variable of interest is endogenous. Given one has a valid instrument that satisfies the conditional moment restriction, many different estimators can be used based on the linear projection of the endogenous variable on the instrument, and transformations of it. I propose an approximate Mean Squared Error (MSE) criterion function to minimize over a set of transformations supplied by the researcher and show it is asymptotically optimal in the sense that the true MSE of the estimator using the optimal number of transformations converges in probability towards the minimum of the true MSE over the set of transformations proposed. In a simulation study, I show the competitive performance of this estimator compared to a variety of estimators used in the literature. I find that it proves particularly competitive when the degree of endogeneity is low, and when the relationship between the endogenous variable and the instrument is highly nonlinear. In other settings, its performance is roughly equivalent to that of the Two Stage Least Squares (2SLS) estimator. In the last chapter, I propose another alternative to instrumental variable estimators that considers the use of kernel based estimators when regressing the endogenous variable on the instruments. I show the resulting estimator is consistent and asymptotically normal, and includes the 2SLS estimator as a special case. Similarly to the second chapter, a simulation study is conducted to show its finite sample behavior.

Document type: 
Thesis
File(s): 
Senior supervisor: 
Bertille Antoine
Department: 
Arts & Social Sciences: Department of Economics
Thesis type: 
(Thesis) Ph.D.

Three essays in applied microeconomics

Author: 
Date created: 
2020-04-02
Abstract: 

In chapter one, Dr. Hendrik Wolff and I examine the effect of an Alabama immigration law on documented immigrants. Alabama's 2011 immigration law (H.B. 56) specifically targets, significantly limits the economic opportunities of, and intensifies the prosecution of undocumented immigrants. Using demographic and citizenship data we test whether the law has led to an unintended reduction of the documented immigrant population of the state. Our hypothesis is that documented immigrants will choose not to locate in Alabama, due to their connections with undocumented immigrants who choose not to live in Alabama because of HB.56. Using synthetic control on data from 2006-2017, we find a significant downward effect on the percentage of the population that is foreign-born non-citizens, and no effect of the law on immigrant citizens. Using data measuring new permanent residents we see no effect on new PRs after the passage of H.B 56. This suggests that the drop in Alabama's immigrant population is likely due to the intended effect of the law discouraging undocumented immigrants from living in Alabama, and that there does not seem to be a similar effect on documented immigrants. In chapter two, Dr. Eric Werker and I estimate community benefits stemming from the signing of benefit sharing agreements associated with two mining projects. Benefit-sharing agreements determine how resource extraction companies and stakeholder communities share the economic rents created by extractive activities. Besides direct financial compensation, BSAs can include preferential access to contracting opportunities for local firms, guarantees of direct employment for local individuals, and other benefits that can be economically quantified. This paper seeks to demonstrate that BSAs can be quantitatively modeled by estimating the expected size of benefits from two BSAs: the Newmont Ahafo gold mine in Ghana, and the Baffinland iron mine in Nunavut, Canada. We compare the levels of expected BSA benefits to a counterfactual scenario in which we imagine the companies carry out their extractive activities in the absence of signing a BSA and estimate the relative contribution from each of financial transfers, jobs, and contracting opportunities. We find that in the Ahafo case the impacted community's discounted benefits from the BSA amount to 1.08\% of the estimated life-of-mine revenue and 2.10\% in the Mary River case, with the primary contributions coming from jobs and financial transfers respectively. Quantifying potential BSA benefits can have practical value for future BSA negotiations and for monitoring the implementation of agreements. In chapter three Dr. Eric Werker and I use a newly created dataset to test hypotheses about what determines the government "take" of gold mining operations worldwide. We define government take as the share of net revenue of a mine collected by the mine's host country government in taxes and other payments. We construct a theoretical model to predict the government take, and then use linear regression to test the agreement between theory and the data. Investment decision theory predicts that governments should decrease their tax rate on mining operations to compensate multinational corporate investors for increased local development costs and political risks. However, higher political risk, and local development requirements are actually associated with higher government take. We find that country-level political economy variables have more predictive power in explaining the patterns determining the government take than the basic investment theory model. We interpret this as evidence that the conventional wisdom surrounding mining investment decisions is incomplete, and that political economy channels may have a role to play in describing the underlying process of determining government take of mining projects.

Document type: 
Thesis
File(s): 
Senior supervisor: 
Hendrik Wolff
Department: 
Arts & Social Sciences: Department of Economics
Thesis type: 
(Thesis) Ph.D.

Translation versus language model pre-training objectives for word sense disambiguation

Date created: 
2019-12-16
Abstract: 

Contextual word representations pre-trained on large text data have advanced state of the art in many tasks in Natural Language Processing. Most recent approaches pre-train such models using a language modeling (LM) objective. In this thesis, we compare and contrast such LM models with the encoder of an encoder-decoder model pre-trained using a machine translation (MT) objective. For certain tasks such as word-sense disambiguation the MT task provides an intuitively better pre-training objective since different senses of a word tend to translate differently into a target language, while word senses might not always need to be distinguished when using an LM objective. Our experimental results on word sense disambiguation provide insight into pre-training objective functions and can help guide future work into large-scale pre-trained models for transfer learning in NLP.

Document type: 
Thesis
File(s): 
Senior supervisor: 
Anoop Sarkar
Department: 
Applied Sciences: School of Computing Science
Thesis type: 
(Thesis) M.Sc.

How’s that sound? Co-designing neurofeedback game audio with children

Date created: 
2020-04-20
Abstract: 

Many children struggle with mental health challenges such as anxiety and attention deficit hyperactivity disorder. Neurofeedback games, like Mind-Full, use portable brain-computer interfaces to help children cope by developing self-regulation skills. These games work by receiving electroencephalographic (EEG) input and relaying that information to users. In theory, feedback may be visual, auditory, tactile, or some combination. In practice, most games are visual. As users gain focus, the game visuals respond on-screen in real time. The reliance on visual feedback can create difficulties in the field—schools are often noisy and filled with distractions, and some children are uncomfortable sitting in silence with adults. Incorporating sound into neurofeedback games could improve usability and, potentially, outcomes. However, there is a basic problem–adult researchers cannot assume to know what sounds might appeal to young users. This prompts important questions: How can children contribute to the co-design of sounds for a neurofeedback system? What sounds do children think would be suitable for each game? In this thesis, I look to children to guide the development and evaluation of sounds for neurofeedback games. I report findings from a co-design study to create sounds for Mind-Full. I worked with 16 children as design partners over five sessions at a school in Vancouver, Canada. I present results from each session, discuss limitations, and review theoretical implications. In this work I find that children can participate in the co-design of sound via ideation, clarification and elaboration of ideas, and evaluation of sounds. The contributions of this work are a set of practical guidelines for co-design with children, an enhanced version of Mind-Full Wind, a summary of the ways in which children can contribute to the co-design of sounds, and insight into the tension between teacher's roles as design partners and facilitators. This work may be helpful for future researchers and designers interested in running co-design studies related to sound.

Document type: 
Thesis
File(s): 
Senior supervisor: 
Alissa N Antle
Department: 
Communication, Art & Technology: School of Interactive Arts and Technology
Thesis type: 
(Thesis) M.Sc.

Zooarchaeology and ethnozoology at Tse’K’wa (Charlie Lake Cave), North Peace Region, British Columbia

Date created: 
2020-05-01
Abstract: 

The archaeological site of Tse’K’wa in the Peace River Region of British Columbia contains faunal remains dating from approximately 11,000 cal BP to the present. Over 20,000 vertebrate faunal remains were analyzed to examine the cultural and ecological record of this site. Of these, 4,000 were identified to the level of Order or more precise classification. Zooarchaeological analyses were integrated with Dane-zaa oral traditions concerning animal species found in the site deposits. This data shows that Ancestral Dene and Dane-zaa community members utilized the site to process and consume animal remains through the entire Holocene sequence. People harvested fish, birds, and mammals from the nearby wetland and forests, and processed and consumed them at Tse’K’wa. There is little evidence for significant environmental change, and there appears to be continuity in the use of sucker, waterfowl, large rodents, snowshoe hare, a wide range of carnivores, and ungulates throughout the site’s history.

Document type: 
Thesis
File(s): 
Senior supervisor: 
Jonathan Driver
Department: 
Environment: Department of Archaeology
Thesis type: 
(Thesis) M.A.

A leap of faith: Motivations for place of worship redevelopment in Vancouver

Author: 
Date created: 
2020-04-20
Abstract: 

This research study explores the motivations for religious congregations in Vancouver, Canada that are redeveloping their sites, building housing (and other ancillary services and uses), while retaining their place of worship function. This is a recent development phenomenon for Vancouver, and at the time of research, no academic studies had yet addressed the topic for the Vancouver context. The purpose is to better understand what internal and external forces may be motivating congregations to pursue such redevelopment schemes, by focusing on their rationale, objectives, and experiences with the projects thus far, from the perspective of the congregations themselves. The lessons learned from this research aim to provide insight on place of worship redevelopment in Vancouver, with a focus on four case studies, and highlight the areas of convergence and divergence from place of worship redevelopment happening in other urban contexts.

Document type: 
Graduating extended essay / Research project
File(s): 
Senior supervisor: 
Peter V. Hall
Department: 
Arts & Social Sciences: Urban Studies Program
Thesis type: 
(Project) M.Urb.

Development through the Indigenous lens – An analysis of First Nations legal frameworks in Canada - AND - Gaming and Indigenous sovereignty discourse – Textual analysis of “invaders” by Elizabeth LaPensée

Author: 
Date created: 
2020-04-23
Abstract: 

Essay 1: Development is the interrelationship between and balance of three pillars namely – Economic Development, Social Development, and Environmental Development to meet the needs of present and future generations. Using the Haudenosaunee Confederacy as a case study, I examined the development philosophies of Indigenous Nations in Canada. I observed that, endogenous development is an intrinsic value of Indigenous growth. However, due to colonization and neo-colonial policies of assimilation, cultural condemnation and land dispossession in contemporary Canada, such growth and development is only possible in sovereign Indigenous Nations. I therefore explored the concept of Indigenous sovereignty in Canada using discourses analysis. By this, I identified key principles of Indigenous sovereignty – liberty, freedom, accountability, collective responsibility, and collective security. This set the framework of analysis of international (the United Nations Declaration on the Rights of Indigenous Peoples (UNDRIP)) as well as domestic policies (Constitution Act 1867, Societies Act of BC) that supports Indigenous sovereignty. Although, there are legal constraints with these legislations in the attainment of sovereignty, the Societies Act offers good grounds to achieve economic sovereignty and sustainable development in the short run. Essay 2: Games facilitate the transfer of knowledge and serve as a medium for knowledge production, memory development and, to some extent, ideology construction. In Indigenous societies, games are played to enhance one’s abilities, stimulate active learning, and reinforce knowledge. Within the digital spheres, games have been designed with algorithms that reinforce these characteristics and, at the same time, foreground dominant ideologies of liberalism, capitalism, and neocolonialism. To counter this tradition and centre minority interests, minoritized game developers re-engineer games to better represent their concerns. This is an interest of Indigenous game design in order to represent Indigenous epistemologies, and tell Indigenous stories using digital technologies, thereby asserting their cultural sovereignty in the digital world. By playing and analyzing the game Invaders developed by Anishinaabe game developer Elizabeth LaPensée in relation to with literature in Indigenous digital studies and gaming, this paper examines how gaming technologies are used to assert Indigenous sovereignty and epistemologies.

Document type: 
Graduating extended essay / Research project
File(s): 
Senior supervisor: 
Karrmen Crey
Department: 
Communication, Art & Technology: School of Communication
Thesis type: 
(Extended Essays) M.A.

Sequence stratigraphy and facies analysis of the viking formation in crossfield and adjacent areas, Alberta, Canada

Author: 
Date created: 
2020-03-18
Abstract: 

The Lower Cretaceous Viking Formation is a siliciclastic unit that accumulated in the Western Interior Seaway and produces hydrocarbons in the Crossfield and surrounding area. The study area encompasses Townships 24–31, Ranges 28W4 to 4W5 in Alberta, Canada. Sedimentological and ichnological observations of 54 cored wells, coupled with the analysis of 1415 geophysical well logs allowed the generation of a high-resolution sequence stratigraphic framework. The Viking Formation is subdivided into four discrete depositional sequences (Sequences 1-4), but this study focuses on Sequence 3, as it constitutes the main cored interval in the Crossfield area. Sequence 3 includes eleven sedimentary facies that are grouped into three facies associations (FA1-3). Deposits of FA1 correspond to the transgressive system tract. The overlying highstand system tract encompasses FA2 and FA3. FA3 contains the reservoir sandstone interval, and comprises sanding-upward successions interpreted to represent deposition in a mixed-process wave-dominated, fluvial-influenced symmetric delta.

Document type: 
Thesis
Senior supervisor: 
James MacEachern
Department: 
Science: Department of Earth Sciences
Thesis type: 
(Thesis) M.Sc.