Waterloo Research

Permanent URI for this collectionhttps://uwspace.uwaterloo.ca/handle/10012/9972

This is the University of Waterloo Research collection.

Waterloo faculty, students, and staff can contact us or visit the UWSpace guide to learn more about depositing their research.

NOTE: This collection does not include graduate Theses & Dissertations or Major Research Projects.

Browse

Now showing 1 - 20 of 3439

Estimating dynamic spillover effects along multiple networks in a linear panel model
(University of Waterloo, 2022-11-17) Possnig, Clemens; Rotarescu, Andreea; Song, Kyungchul
Spillover of economic outcomes often arises over multiple networks, and distinguishing their separate roles is important in empirical research. For example, the direction of spillover between two groups (such as banks and industrial sectors linked in a bipartite graph) has important economic implications, and a researcher may want to learn which direction is supported in the data. For this, we need to have an empirical methodology that allows for both directions of spillover simultaneously. In this paper, we develop a dynamic linear panel model and asymptotic inference with large n and small T, where both directions of spillover are accommodated through multiple networks. Using the methodology developed here, we perform an empirical study of spillovers between bank weakness and zombie-firm congestion in industrial sectors, using firm-bank matched data from Spain between 2005 and 2012. Overall, we find that there is positive spillover in both directions between banks and sectors.
Descriptive labour market outcomes of immigrant women across Europe
(University of Waterloo, 2022) Adsera, Alicia; Ferrer, Ana; Herranz, Virginia
We consider the job progression of immigrant women in five European countries: France, Italy, Spain, Sweden and the UK. We complement data from the European Labour Force Survey (2005-2015), with information about the skills contained in the jobs held on by women, using data from the O*Net. In particular, we focus on analytical and strength skills in immigrant's jobs and compare them to those required by jobs held by similar native women. Even though immigrants experience upon arrival a gap in participation relative to the native born, they gradually increase participation during the first ten years spent in the country (approximately, 1% per year in Spain, Italy and the UK, and 2% and 4% per year in France and Sweden respectively). Our results reveal significant differences across countries of origin as well as differences within countries over the period of analysis. Recent immigrant women show relatively large gaps in the analytical skill content of the jobs they held relative to native-born women across our host countries. Further, with the exception of immigrants to pain, they also work jobs with higher requirements of strength than their native-born counterparts do. Although educated immigrants show a different pattern in most countries (included Spain). We find differences within countries over the period of analysis that may be consistent with the variation of incentives to move depending on the business cycle at arrival - particularly given the meager opportunities in many destination countries during aftermath of the recent great recession.
Immigrant gaps in parental time investments into children's human capital activities
(University of Waterloo, 2022) Ferrer, Ana; Mascella, Allison
Current and future well-being and economic prosperity of children depend in large part on the nuances of decisions made by parents with respect to familial resources, an important part of which regard the time spent in the company of children. We estimate differences in the time that immigrant and Canadian-born parents allocate to child-care activities relative to other activities using the time diaries from the General Social Survey. We find that mothers born abroad spend more time at work and less time in leisure but there is no significant difference in time devoted to household production or child service between them and Canadian-born mothers. Despite not finding differences by immigration status in the total care-time parents provide for their children, we do find significant differences - by immigrant status - in time specifically devoted to human capital investment activities with children: African, Asian, European and South-Central American mothers spend up to 30 more minutes daily in these activities than the Canadian born. We further assess the patterns of time use of second-generation young adults and find that they spend more time on education and homework compared to third generation or higher young adults. This supports a plausible effect of the time invested in children's human capital generating activities by immigrant parents on their Canadian-born children.
The work trajectories of married Canadian immigrant women, 2006-2019
(University of Waterloo, 2022) Ferrer, Ana; Pan, Annie; Schirle, Tammy
The behaviour of married immigrant women regarding fertility and labour markets is an essential piece to understand the economic and cultural integration of immigrant households. However, the contribution of married immigrant women to the Canadian labour market was - until recently - considered of secondary importance and their labour market choices studied within a significant fraction of married immigrant women make labor supply decisions (and face barriers) similar to those of native-born married women. We show that this is the case in Canada as well, by estimating the progress of immigrant women over the 2000s. We use traditional measures of labour market attachment, such as participation, employment and wages, but also novel measures of labour market dynamics, such as transitions across labour market states. Differences in transition rates can reveal higher fragility of work for immigrant women, or reveal the extent to which immigrant women respond to family income shocks - the added worker effect. Results show that immigrant women are less likely to transition into employment - more likely to transition out of employment to either unemployment or inactivity - and more likely to respond to income shocks than the Canadian born. There is evidence of a gradual convergence with years spent in Canada to the outcomes of the Canadian born, which is much slower for immigrant women than immigrant men.
Wild boostrap inference for penalized quantile regression for longitudinal data
(University of Waterloo, 2022-10-18) Lamarche, Carlos; Parker, Thomas
The existing theory of penalized quantile regression for longitudinal data has focused primarily on point estimation. In this work, we investigate statistical inference. We propose a wild residual bootstrap procedure and show that it is asymptotically valid for approximating the distribution of the penalized estimator. The model puts no restrictions on individual effects, and the estimator achieves consistency by letting the shrinkage decay in importance asymptotically. The new method is easy to implement and simulation studies show that it has accurate small sample behavior in comparison with existing procedures. Finally, we illustrate the new approach using U.S. Census data to estimate a model that includes more than eighty thousand parameters.
Uniform inference for value functions
(University of Waterloo, 2022-11-10) Firpo, Sergio; Galvao, Antonio F.; Parker, Thomas
We propose a method to conduct uniform inference for the (optimal) value function, that is, the function that results from optimizing an objective function marginally over one of its arguments. Marginal optimization is not Hadamard differentiable (that is, compactly differentiable) as a map between the spaces of objective and value functions, which is problematic because standard inference methods for nonlinear maps usually rely on Hadamard differentiability. However, we show that the map from objective function to an Lp functional of a value function, for 1 ≤ p ≤ ∞, are Hadamard directionally differentiable. As a result, we establish consistency and weak convergence of nonparametric plug-in estimates of Cramer-von Mises and Kolmogorov-Smirnov test statistics applied to value functions. For practical inference, we develop detailed resampling techniques that combine a bootstrap procedure with estimates of the directional derivatives. In addition, we establish local size control of tests which use the resampling procedure. Monte Carlo simulations assess the finite-sample properties of the proposed methods and show accurate empirical size and nontrivial power of the procedures. Finally, we apply our methods to the evaluation of a job training program using bounds for the distribution function of treatment effects.
Fighting for fares: Uber and the declining market price of licensed taxicabs
(University of Waterloo, 2022-04-22) Garnham, Alina; Stacey, Derek
In this paper, we study how the emergence of Uber in a large North American city affects the financial value of taxicab licenses. A taxicab license provides a claim to a stream of dividends in the form of rents generated by operating the taxicab or leasing the license. The introduction of Uber undoubtedly affects the anticipated stream of dividends because Uber drivers capture part of the farebox revenue that might otherwise go to the owners/drivers of licensed taxicabs. At the same time, the launch of Uber's innovative technology-driven approach to the provision of ride-hailing services can be viewed as a partial obsolescence of the traditional taxicab approach. The economic incentives facing market participants may therefore change as Uber gains momentum in the ride-hailing market, which could further affect the market value of licensed taxicabs. Using transaction-level data, we apply a theory of asset pricing to the secondary market for Toronto taxicab licenses to explore these potential price effects. We learn that both the farebox and innovation effects contribute to the overall decline in market value, with the farebox effect account for just over half of the $170K price decline from 2011 to 2017. We explore the welfare implications for taxicab license owners with counterfactual simulations. We find that, consistent with the anti-Uber protests organized by Toronto taxi drivers, there was a high willingness to pay among license holders to prevent or postpone the launch of Uber's ridesharing services.
Learning new bias: Misspecifications and consequences
(University of Waterloo, 2023-09) Hu, Lin; Kovach, Matthew; Li, Anqi
We study how a decision maker (DM) learns about the bias of unfamiliar news sources. Absent any frictions, a rational DM uses known sources as a yardstick to discern the true bias of a source. If a DM has misspecified beliefs, this process fails. We derive long-run beliefs, behavior, welfare, and corresponding comparative statistics, when the DM has dogmatic, incorrect beliefs about the bias of known sources. The distortion due to misspecified learning is succinctly captured by a single-dimensional metric we introduce. Our model generates the hostile media effect and false polarization, and has implications for fact-checking and misperception recalibration.
Rationally inattentive statistical discrimination: Arrow meets phelps
(University of Waterloo, 2023-08) Echenique, Federico; Li, Anqi
When information acquisition is costly but flexible, a principal may rationally acquire information that favors "majorities" over "minorities" unless the latter are strictly more productive than the former. Majorities therefore face incentives to invest in becoming productive, whereas minorities are discouraged from such investments. The principal, in turn, focuses scarce attentional resources on majorities precisely because they are likely to invest. We give conditions under which the resulting discriminatory equilibrium is most preferred by the principal, despite that all groups are ex-ante identical. Our results add to the discussions of affirmative action, implicit bias, and occupational segregation and stereotypes.
Learning to best reply: On the consistency of multi-agent reinforcement learning
(University of Waterloo, 2023-08-10) Possnig, Clemens
This paper provides asymptotic results for a class of model-free actor-critic batch - reinforcement learning algorithms in the multi-agent setting. At each period, each agent faces an estimation problem (the critic, e.g. a value function), and a policy updating problem. The estimation step is done by parametric function estimation based on a bath of past observations. Agents have no knowledge of each others incentives and policies. I provide sufficient conditions for each agent's parametric function estimator to be consistent in the multi-agent environment, which enables agents to learn to best respond despite the non-stationarity inherent in multi-agent systems. The conditions depend on the environment, batch size, and policy step size. These sufficient conditions are useful in the asymptotic analysis of multi-agent learning, e.g. in the application of long-run characterisations using stochastic approximation techniques.
Reinforcement learning and collusion
(University of Waterloo, 2023-07-24) Possnig, Clemens
This paper presents an analytical characterization of the long rung policies learned by algorithms that interact repeatedly. These algorithms update policies which are maps from observed states to actions. I show that the long rung policies correspond to equilibria that are stable points of a tractable differential equation. As a running example, I consider a repeated Cournot game of quantity competition, for which learning the stage game Nash equilibrium serves as a non-collusive benchmark. I give necessary and sufficient conditions for this Nash equilibrium not to be learned. These are requirements on the state variables algorithms use to determine their actions, and on the stage game. When algorithms determine actions based only on the past period's price, the Nash equilibrium can be learned. However, agents may condition their actions on richer types of information beyond the past period's price. In that case, I give sufficient conditions such that the policies coverage with positive probability to a collusive equilibrium, while never converging to the Nash equilibrium.
A rational inattention theory of echo chamber
(University of Waterloo, 2023-07) Hu, Lin; Li, Anqi; Tan, Xu
We develop a rational inattention theory of echo chamber, whereby players gather information about an uncertain state by allocating limited attention capacities across biased primary sources and the other players. The resulting Poisson attention network transmits information from the primary source to a player either directly or indirectly through the other players. Rational inattention generates heterogeneous demands for information among players who are initially biased towards different decisions. In an echo chamber equilibrium, each player restricts attention to his own-biased source and like-minded friends, as the latter attend to the same primary source as his, and so could serve as secondary sources in case the information transmission from the primary source to him is disrupted. We provide sufficient conditions that give rise to echo chamber equilibria, characterize the attention networks within echo chambers, and use our results to inform the design and regulation of information platforms.
Head Kinematic Measurements and Finite Element Modeling of Canadian Armed Forces Operators Firing Three Long-Range Rifle Configurations
(2026-06-12) Seeburrun, Tanvi; Hartlen, Devon C.; Bustamante, Michael B.; St-Onge, Gabriel; Ouellet, Simon; Cronin, Duane S.
Mild traumatic brain injury (mTBI) symptoms have been associated with repeated exposure to the recoil of long-range rifles. However, there is limited physical data on head responses to rifle recoil and no consistent approach to quantitatively compare rifle configurations that may mitigate head response to recoil. In this study, the head kinematics of Canadian Armed Forces volunteers firing long-range rifles were measured and used as input to a finite element (FE) head model, enabling comparisons across different operators and rifle configurations. Head kinematics were measured with instrumented mouthguards for three rifle configurations: a 0.50 caliber rifle, a 0.338 caliber rifle, and a 0.338 caliber rifle with a recoil mitigation system (RMS). Measured head kinematics were used as input loading conditions to an FE head model to calculate brain tissue strains resulting from recoil, which were quantified using cumulative strain volume (CSV) curves. It was found that the 0.50 caliber rifle induced significantly higher strains than the 0.338 caliber rifle, while the RMS system reduced brain strain for the 0.338 caliber rifle. Characteristics such as differing anthropometrics, posture, or technique may influence brain strains, explaining the differences between volunteers. Isolating aspects of head kinematics, specifically rotation in the sagittal plane, identified it as having the largest contribution to brain strain. The findings from this study provide foundational data on the magnitudes of head kinematics experienced by volunteers when firing long-range rifles and serve as an important step toward mitigation of recoil exposures.
Indigenous Data Sovereignty Primers: A Guide for Researchers
(University of Waterloo, 2026-06) Indigenous Sovereignty, Autonomy, Governance, Ethics (ISAGE); Hill, Jade; Jamieson, Jordan; Anderson, Sara; University of Waterloo, RDM Institutional Strategy Working Group; University of Waterloo, RDM Advisory Group
Research does not happen in a vacuum. It is shaped by histories, relationships, and systems of power that continue to influence how knowledge is produced, shared, and valued. For Indigenous Peoples, research and data practices have been deeply entangled with colonial systems that have extracted knowledge, resources, and stories without consent, without context, and without accountability. For generations, data about Indigenous Peoples, lands, and knowledge systems has been collected by governments, academic institutions, and other external actors and used in ways that have defined, categorized, and governed Indigenous lives. Census data has been used to control identity. Environmental data has enabled resource extraction on Indigenous territories. Cultural knowledge has been recorded, archived, and circulated far from the communities it comes from. In many cases, these processes have removed knowledge from its relational and cultural context, turning living systems of knowing into objects of study. These practices are not only in the past. They continue, often in less visible ways. Data may be collected under the assumption that it is neutral, publicly available, or disconnected from the people and places it relates to. Research may proceed without recognizing Indigenous presence, authority, or governance. Even well-intentioned work can reproduce extractive patterns when it does not account for the histories and relationships that shape Indigenous data. Indigenous Data Sovereignty (IDSov) emerges within this context as both a response and a reassertion. It reflects the inherent rights of Indigenous Peoples to govern data about their communities, lands, cultures, and knowledge systems. It is rooted in self-determination, but it is also grounded in longstanding Indigenous laws, governance structures, and relationships that have always guided how knowledge is held, shared, and protected. IDSov challenges the assumption that data can be separated from people and place, and instead affirms that data is relational, contextual, and governed. This guide was developed to support researchers in engaging with these realities. It is not simply a technical resource about data management, but an invitation to rethink how research is approached, particularly in relation to Indigenous Peoples. Many researchers may not set out to work with Indigenous communities but may still encounter Indigenous data through land-based research, demographic datasets, archival materials, or emerging technologies. In these situations, responsibilities do not disappear because relationships are less visible. The Indigenous Data Sovereignty Primers provide accessible, practical guidance across research contexts. Each primer can be read on its own, but together they offer a broader understanding of how Indigenous Data Sovereignty applies across the research lifecycle, from data collection to analysis and sharing. This guide does not offer a single prescriptive approach. Indigenous governance systems and community expectations are diverse and context specific. Instead, it encourages researchers to ask critical questions about authority, relationships, and responsibility, and to move away from extractive approaches toward practices grounded in respect, reciprocity, and accountability.
Environmental bonds and public liability for resource extraction site cleanup
(University of Waterloo, 2024) Insley, Margaret
Governments have been left with large liabilities for cleanup at natural resource extraction sites after firms have declared bankruptcy. This research studies the impact of different forms of financial assurance on a firm's optimal actions over the full life cycle of a hypothetical natural gas well, in a world of uncertain natural gas prices, when firm bankruptcy may shift cleanup costs to the government. A firm's stochastic optimal control problem is described by an HJB equation with the natural gas price modelled as a stochastic differential equation. The impact of financial assurance is examined in relation to firm investment decisions, the cleanup liability imposed on government and resource taxation revenue. A Cash Deposit and Surety Bond are contrasted with the case of no financial assurance requirement. The "fair" fee (assuming the absence of arbitrage opportunities) is determined for the Surety Bond issued by a third party. Numerical results demonstrate that in the presence of distortionary taxes, there is a trade off between indemnifying the government against cleanup costs versus maintaining government tax and royalty revenues. A numerically plausible case is presented in which the total value of the natural gas well (to the firm and the government) is not increased by the imposition of a strict form of financial surety.
Canada's racialized immigrant women
(University of Waterloo, 2024) Ferrer, Ana; Dhatt, Sumeet Singh
Immigrants have traditionally lagged behind labour outcomes of Canadian born workers, a fact that is more obvious for immigrant women and for recent arrivals (those entering Canada within the last five years). In this report we explore the barriers and challenges faced by racialized newcomer women in the Canadian labour market and how differences in their characteristics are (or aren't) related to differences in labour market outcomes. We use a specially designed survey to capture the experiences of a sample of racialized newcomer women regarding integration into the labour market and what resources and strategies have been most helpful in achieving career success and improving their quality of life. We follow with an in-depth analysis of the labour market environment of immigrant women to Canada using data from the Labour Force Survey and the O*Net database. This allows us to quantify to what extent immigrant women may be facing barriers and challenges in the labour market, not only along many standard measures of job quality, such as employment, pay, or type or contract, but also examining other non-standard measures of job quality that are informative of the resilience of the jobs immigrants hold, such as the tasks they perform in their jobs.
The economics of Canadian immigration levels
(University of Waterloo, 2024) Doyle, Matthew; Skuterud, Mikal; Worswick, Christopher
In the hope of addressing chronic labour shortages and sluggish economic growth, the Canadian government plans to increase immigration in the coming years to per capita levels not reached since the 1920s. We argue that economic immigration in the Canadian context should aim to boost GDP per capita in the full population including the newcomers. We then examine the potential for increases in Canadian immigration levels to achieve this objective. Our analysis suggests that Canada is not well-positioned to leverage heightened immigration to boost GDP per capita owing primarily to weak capital investment and quantity-quality tradeoffs in immigrant selection. We conclude by providing a framework for identifying the optimal level of economic immigration.
Establishing a FAIR, CARE, and efficient synthetic health data sharing ecosystem for Canada
(University of Waterloo, 2024) Chen, Helen; Grossman, Maura R.; Sen, Anindya; Tsao, Shu-Feng
Obtaining access to real-world health data is a significant challenge, mainly due to privacy and security implications. Consequently, researchers and technology innovators - particularly those operating in the health data science and AI technology development spaces - increasingly resort to synthetic health data to bridge the data gap. High-quality synthetic data has the potential to expedite research and development of novel technologies. However, synthetic health datasets in Canada are scarce, and no existing synthetic health datasets conform to the Findable, Accessible, Interoperable, and Reusable (FAIR) standards. Moreover, while federated learning with synthetic health data. This paper explores the ethical considerations and value proposition of generating and sharing synthetic health data. Our goal is to facilitate the development of a reliable and sustainable synthetic data infrastructure that supports the ethical, responsible, and efficient use of synthetic health data. An important contribution of this research is the establishment of a framework that balances the social benefits of innovation from data sharing with the social costs that occur when individual privacy is compromised. The use of synthetic data significantly reduces the potential for individual harm and is a cost-effective means to lower data-sharing costs. We believe that this framework will pave the way for a more robust and secure synthetic data ecosystem, enabling the generation of valuable insights that can drive positive health outcomes for Canadians.
Implicit incentive provision with misspecified learning
(University of Waterloo, 2025) Echenique, Federico; Li, Anqi
We study misspecified Bayesian learning in principal-agent relationships, where an agent is assessed by an evaluator and rewarded by the market. The agent's outcome depends on their innate ability, costly effort - whose effectiveness is governed by a productivity parameter - and noise. The market infers the agent's ability from observed outcomes and rewards them accordingly. The evaluator conducts costly assessments to reduce outcome noise, which shapes the market's inferences and provide implicit incentives for effort. Society - including the evaluator and the market - holds dogmatic, inaccurate beliefs about ability, which distort learning about effort productivity and effort choice. This, in turn, shapes the evaluator's choice of assessment. We describe a feedback loop linking misspecified ability, biased learning about effort, and distorted assessment. We characterize outcomes that arise in stable steady states and analyze their robust comparative statistics and learning foundations. Applications to education and labor market reveal how stereotypes can reinforce across domains - sometimes disguised as narrowing or even reversals of outcome gaps - and how policy interventions targeting assessment can help.
The impact of infection on population health: Results of the Ontario burden of infectious diseases study
(Public Library of Science, 2012-09-04) Kwong, Jeffrey C.; Ratnasingham, Sujitha; Campitelli, Michael A.; Daneman, Nick; Deeks, Shelley L.; Manuel, Douglas G.; Allen, Vanessa G.; Bayoumi, Ahmed M.; Fazil, Aamir; Fisman, David N.; Gershon, Andrea S.; Gournis, Effie; Heathcote, E. Jenny; Jamieson, Frances B.; Jha, Prabhat; Khan, Kamran M.; Majowicz, Shannon E.; Mazzulli, Tony; McGeer, Allison J.; Muller, Matthew P.; Raut, Abhishek; Rea, Elizabeth; Remis, Robert S.; Shahin, Rita; Wright, Alissa J.; Zagorski, Brandon; Crowcroft, Natasha S.
Background Evidence-based priority setting is increasingly important for rationally distributing scarce health resources and for guiding future health research. We sought to quantify the contribution of a wide range of infectious diseases to the overall infectious disease burden in a high-income setting. Methodology/Principal Findings We used health-adjusted life years (HALYs), a composite measure comprising premature mortality and reduced functioning due to disease, to estimate the burden of 51 infectious diseases and associated syndromes in Ontario using 2005–2007 data. Deaths were estimated from vital statistics data and disease incidence was estimated from reportable disease, healthcare utilization, and cancer registry data, supplemented by local modeling studies and national and international epidemiologic studies. The 51 infectious agents and associated syndromes accounted for 729 lost HALYs, 44.2 deaths, and 58,987 incident cases per 100,000 population annually. The most burdensome infectious agents were: hepatitis C virus, Streptococcus pneumoniae, Escherichia coli, human papillomavirus, hepatitis B virus, human immunodeficiency virus, Staphylococcus aureus, influenza virus, Clostridium difficile, and rhinovirus. The top five, ten, and 20 pathogens accounted for 46%, 67%, and 75% of the total infectious disease burden, respectively. Marked sex-specific differences in disease burden were observed for some pathogens. The main limitations of this study were the exclusion of certain infectious diseases due to data availability issues, not considering the impact of co-infections and co-morbidity, and the inability to assess the burden of milder infections that do not result in healthcare utilization. Conclusions/Significance Infectious diseases continue to cause a substantial health burden in high-income settings such as Ontario. Most of this burden is attributable to a relatively small number of infectious agents, for which many effective interventions have been previously identified. Therefore, these findings should be used to guide public health policy, planning, and research.

Browse

Recent Submissions