Main Content

TD MDAL Research Grant - 2022 awards

Grant recipients

Data analytics is relevant across a wide range of areas, applications from faculty and PhD students from all areas of the Graduate Department of Management were considered. 

Special thanks to two new TD MDAL faculty fellows, Azarakhsh Malekian (OM) and Mitch Hoffman (Strategy) who helped adjudicate the grant proposals. Both are stellar researchers in the theory and application of data analytics and we are excited to support their future research.

The lab awarded funds to the successful projects listed below in 2022. 

Faculty and PhD Student recipients include:




Angele Beausoleil, Assistant Professor of Business Design and Innovation, “Measuring and Mining Innovativeness at TD Bank. (with Anjana Dattani, Senior Research Associate)


Julieta Caunedo, Assistant Professor, Economics Analysis and Policy, “Optimizing rental equipment services in space.”

Elizabeth DhueyAssociate Professor of Economic Analysis and Policy, “Using AI and Machine Learning for Meta-Analysis in the Social Sciences.” (with Michal Perlman, Professor of Applied Psychology and Human Development; Linda White, RBC Chair in Economic and Public Policy and a Professor of Political Science)

Charles Martineau, Assistant Professor of Finance (UTSC), “Getting Advice from Retail Investors.(with Edna Lopez Avila, PhD student in Finance)

Chay Ornthanalai, Associate Professor of Finance, Equity Option Trading on Newly Public Firms

Rachel Ruttan, Assistant Professor of Organizational Behavior and Human Resources, “A ‘Big Data’ Investigation of Deviant Communication Strategies.” (with Katy DeCelles, Professor of Organizational Behavior; Jun Lin Postdoctoral Associate, New York University)

Jincheng Tong, Assistant Professor of Finance (UTSC), “Data, Corporate policies, and Asset Prices.” 

PhD students

Andy Back, PhD Student, Strategic Management, “One-sided Narratives of the Strategic Alliances: How Do Firms Strategically Narrate Their Alliance Formation?” (with Bill McEvily Professor of Strategic Management; Navid Asgari, Professor of Corporate Strategy, Fordham University)

Kane Bae, PhD Student, Finance, Inventory Investment and Firm Risk.” (with Alexandre Corhay, Assistant Professor of Finance; Jincheng Tong, Assistant Professor of Finance UTSC)

Noémie Bucourt, PhD Student, Finance, “Public and private financing complementarities in financing climate transition.

Manuela Collis, PhD Student, Strategic Management, “Gender gaps in knowledge contribution and patenting.” (with Nicola Lacetera, Professor in the Department of Management)

Sepideh Hosseini, PhD Student, Economic Analysis & Policy, “Allow Them to Track You? Evidence from Apple’s Application Tracking Transparency Policy.”

Saman Lagzi, PhD Student, Operations Management, “Use Neural Networks to Guide Data-driven Operational Decisions.” (with Joseph Milner, Professor, Operations Management and Statistics area, Magna Professorship in Management, Vice-Dean, MBA Programs; Ningyuan Chen, Assistant Professor (Business Analytics) Management)

Peng Liu, PhD Student, Finance, “A probabilistic view on understanding time-varying risk premium.” (with Bing Han, Professor of Finance, TMX Chair in Capital Markets; Boyu Wang, Assistant Professor, Department of Computer Science, University of Western)

Junhao Liu, PhD Student, Accounting, “Federal Open Market Committee Meetings and Analyst Forecasts.” (with Ole-Kristian Hope, Deloitte Professor of Accounting; ; Junhao Liu, PhD student, Accounting; Mingyue Zhang, PhD student, Accounting)

Y. Christine Liu, PhD Student, Accounting, “Beyond the Twilight Zone: Restructuring and the Resurrection of Zombie Firms.

Xijiang Su, PhD Student, Accounting, “Strategic Disclosure of Mutual Fund Performance.” (with Ole-Kristian Hope, Deloitte Professor of Accounting)


Abstracts provided below:

“Measuring and Mining Innovativeness at TD Bank.” 

Angele Beausoleil, Assistant Professor of Business Design and Innovation, (with Anjana Dattani, Senior Research Associate)

Abstract: In an era of growing uncertainty, organizations face the dilemma of just-in-time innovation to survive, and the associated risks (and costs) to infuse a continuous innovation culture. To better navigate this tension, firms need to develop, attract and retain leaders who are strategic, customer-centred, creative and agile, or in other words, innovative. Innovative leaders demonstrate a high level of innovativeness or innovative capacity. Innovativeness is described as the ability and willingness to adopt new ideas, think creatively and critically, act with curiosity and empathy, and tolerate uncertainty through a business transformation or innovation process. Who and where are these innovative leaders inside organizations? To discover these individuals, a new innovativeness instrument is in development that measures and mines data patterns of innovative capacity indicators. This instrument combines three scientific scales that determine an individual’s rate of innovation adoption, core personality traits and adaptability dimensions. The beta version has been tested by TD Bank’s Business Strategy Group ‘high potentials’ over the past three years. This research grant will support new resources and new data analytics tools for improved pattern recognition, data visualization and innovativeness research insights.


“Optimizing rental equipment services in space.”

Julieta Caunedo, Assistant Professor, Economics Analysis and Policy

Abstract: We have partnered up with one of the largest equipment rental providers in the world to optimize clustering of demand and service delivery in space and time. This proposal will found the development of an algorithm to link information gathered at different points of the service delivery, i.e. booking requests with the service providers, to be able to better exploit demand and supply patterns. We will also develop an index of booking agents' ability to cluster demand in space and time and suggestions for further clustering and its direction.


“Using AI and Machine Learning for Meta-Analysis in the Social Sciences.”

Elizabeth Dhuey, Associate Professor of Economic Analysis and Policy (with Michal Perlman, Professor of Applied Psychology and Human Development; Linda White, RBC Chair in Economic and Public Policy and a Professor of Political Science)

Abstract: Meta-analysis and systematic reviews require significant manpower to complete in the social sciences. This is due to the lack of shared vocabulary across disciplines and topics. Therefore, when searching for relevant manuscripts, the researchers often have thousands of manuscript abstracts to hand code for relevance. This project will assess whether using AI can aid social science researchers in creating a more efficient meta-analysis/systematic review process by decreasing the number of abstracts that need to be hand-coded.


“Getting Advice from Retail Investors.” 

Charles Martineau, Assistant Professor of Finance, UTSC (with Edna Lopez Avila, PhD student in Finance)

Abstract: Retail investors have access to a growing number of platforms conveying investment advice. Is such advice beneficial to market efficiency? WallStreetBets and Seeking Alpha are popular investment advice platforms for retail traders with different barriers to information production. In contrast to WallStreetBets, investments posts on Seeking Alpha are reviewed by moderators and generate revenue for authors. These barriers influence the timeliness of information production. Focusing on earnings announcements, WallStreetBets produces more timely information ahead of announcements than Seeking Alpha and affects market efficiency in two ways: investment advice benefits price efficiency in days leading to positive earnings news but worsens price efficiency before negative earnings news. Such asymmetric price adjustment is caused by over-optimistic advice found on WSB and SA. Investment advice from more sophisticated advisors, i.e., financial analysts, does not display such over-optimism.


“Equity Option Trading on Newly Public Firms.”

Chay Ornthanalai, Associate Professor of Finance 

Abstract: We document strong evidence for the negative and persistent impact of option listing on stock returns of newly public firms. The result is not explained by exchanges' timing of when to list options and is specific to newly public firms. Option listing that is not accompanied by the start of option trading or listing on seasoned firms do not yield the same effect. We rule out the relaxation of short-sale constraint as an explanation. Using data from the security-lending market, we find that it is more difficult and more costly to short shares after option is listed suggesting that short-sale constraint worsens. Directional option volume sourced by investor types shows that after option listing, firm proprietary traders accumulate large negative exposure on newly public shares by buying puts and writing calls while public customers trade in the opposite direction. Negative exposure held by proprietary traders in option contracts strongly predicts lower stock returns and explains the increased short-selling demand and cost. Our results highlight the role of option market as a venue for information-based trading rather than alleviating short-sale constraint.


“A ‘Big Data’ Investigation of Deviant Communication Strategies.” 

Rachel Ruttan, Assistant Professor of Organizational Behavior and Human Resources (with Katy DeCelles, Professor of Organizational Behavior; Jun Lin Postdoctoral Associate, New York University)

Abstract: One common strategy to negate deviant communications online, such as hate speech, has involved regulations and policing, including the use of bans. For instance, Reddit closed several subreddits associated with hate speech, and based on existing research, these bans were deemed successful in reducing hate speech on the platform (e.g., Chandrasekharan et al., 2017). However, existing work examined the use of explicit language before-and-after bans or regulations were put into place. We propose that, in the face of regulation, rather than being compliant, deviant communicators instead shift communication strategies, adopting more symbolic (thus containing a lower risk of detection) communication styles that are still successful in finding likeminded users. We will examine this question using two real-world data sets. The first follows users of the banned threads on Reddit from an archive and the second will involve collecting tweets (estimated more than 10 million tweets covering nearly 10 years) through the Twitter official API.


“Data, Corporate policies, and Asset Prices.” 

Jincheng Tong, Assistant Professor of Finance, UTSC

Abstract: While the data economy has altered the way consumers shop and businesses operate, it has only recently permeated researchers' thinking about the aggregate economy and financial markets. Macroeconomists have developed comprehensive theoretical frameworks to better understand the interaction between data growth and traditional macroeconomic issues such as firm dynamics, GDP measurement, etc. However, it may surprise you to learn how little is known about the effect of data on firms' optimal corporate decisions, asset returns, etc. We fill this gap in the literature with this project by examining the growth of the data economy and its implications for corporate policies and equilibrium stock returns in the cross-section of publicly traded firms.


“One-sided Narratives of the Strategic Alliances: How Do Firms Strategically Narrate Their Alliance Formation?” 

Andy Back, PhD Student, Strategic Management (with Bill McEvily Professor of Strategic Management; Navid Asgari, Professor of Corporate Strategy, Fordham University)

Abstract: This project attempts to push network research forward by studying network content, especially how the content flowing through network ties is narrated by firms. To gain insight into how firms strategically narrate their network ties, I consider a particularly important type of network ties, strategic alliances, and focus on instances of disclosure by only one of the firms involved. To gain insight into firms’ narration activities, I exploit the relative status of alliance partners as the main explanatory variable of interest. The research question I address is “to what extent does a firms’ own status and their partners’ status affect how firms narrate their alliances?” My project studies two narratives of strategic alliances: alliances narrated 1) only by low-status firms and 2) only by high-status firms. To address the research question, I intend to adopt a novel approach to analyzing textual content using natural language processing (NLP).


“Inventory Investment and Firm Risk.” 

Kane Bae, PhD Student, Finance (with Alexandre Corhay, Assistant Professor of Finance; Jincheng Tong, Assistant Professor of Finance UTSC)

Abstract: Soon after the pandemic arose, firms experienced frequent stock-outs and supply-chain disruptions. Thus, they are paying more attention to inventories available for sale. This research project asks how inventory investment affects firms’ risk in cross-section. This project plans to pursue this question both empirically and theoretically. First, using accounting variables and text data from companies’ annual reports, this research will establish solid relationships between a firms risk premium and inventory-sales ratio data. Moreover, this project will develop a theoretical model that incorporates realistic elements where firms cannot instantly choose production and must prepare inventories to meet uncertain demand. Given recent supply chain disruptions and soaring inflation due to a shortage of goods, I expect this research will draw significant attention from policy makers and researchers. The TD Management Data and Analytics Lab funding is necessary for the acquisition of research assistance to implement this research project and for its success.


“Public and private financing complementarities in financing climate transition.”

Noémie Bucourt, PhD Student, Finance

Abstract: I intend to study interactions between governmental financial support to companies and allocation of financial capital when financing climate transition. "Green" technologies that will allow society to pollute less generate large, long-term, positive externalities nearly no private investor wants to pay for unless governments set the right incentives. However, the latter are not neutral nor informed and one concern is the way they choose which technology to foster. Private investors, needed to financially pay for climate transition, may therefore fund technologies that are supported by governments but not efficient. How do investors react to public intervention? I would like to look at auctions that are organized by governments when they want to produce renewable energy and use the fact that some auctions do not specify the technology that can be used by the company while some others do.


“Gender gaps in knowledge contribution and patenting.”

Manuela Collis, PhD Student, Strategic Management (with Nicola Lacetera, Professor in the Department of Management)

Abstract: Innovation is considered the main driver of economic growth and the biggest contributor to societal progress. As such, scientists have long studied the conditions under which innovation - and creative ideas more generally - emerge. A growing body of evidence suggests that minority group members or diverse teams have the potential to produce the most innovative ideas. However, diversity, or lack thereof, is currently a problem in settings where ideas matter. With this work, I explore gender differences in a particular setting of idea production: academic publications. That is, I analyze the role of gender in a scientist’s decision to contribute their knowledge and decision what topic they want to contribute their knowledge to. This analysis will be extended to patenting activities which will allow us to uncover whether a scientist's knowledge contributions are predictive of their patenting activities.


“Allow Them to Track You? Evidence from Apple’s Application Tracking Transparency Policy.”

Sepideh Hosseini, PhD Student, Economic Analysis & Policy

Abstract: I investigate how privacy protection rules affect privacy choices of application publishers in the context of their interaction with third-party trackers. As a natural experiment, I use a change in Apple's privacy policy known as App Tracking Transparency, which was implemented in April 2021. As per this policy, application developers must explicitly request permission to follow users across applications and websites. Consequently, third-party trackers, particularly advertisers, are predicted to be negatively impacted. I also expect that publishers with an ad-based income model, which are primarily composed of free app creators, to be most affected. I empirically examine the effect, using the data from Apple’s App Store and Google’s Play Store top chart applications rankings before and after the policy implementation. I intend to find out whether the app publishers are now generally more privacy conscious. I also intend to explore the types of app publishers that are more susceptible to this policy change.


“Use Neural Networks to Guide Data-driven Operational Decisions.” 

Saman Lagzi, PhD Student, Operations Management (with Joseph Milner, Professor, Operations Management and Statistics area, Magna Professorship in Management, Vice-Dean, MBA Programs; Ningyuan Chen, Assistant Professor (Business Analytics) Management)

Abstract: We use Neural Networks as approximation algorithms to address complex problems. Given sample data that includes observations of decision variables, covariates and the objective function value, we train a neural network, where the input of the neural network is the covariates and decision variables, and its output is the predicted value for the objective function. Then, for a given set of covariates, the decision variable is chosen so the predicted value of the neural network is optimal. We characterize the performance of our methodology in terms of the generalization bound of the Neural Network and show strong performance on both the Newsvendor problem and the assortment pricing problem.

The full paper is under review at Management Science, the link to access the paper is provided here.


“Federal Open Market Committee Meetings and Analyst Forecasts.”

Junhao Liu, PhD Student, Accounting (with Ole-Kristian Hope, Deloitte Professor of Accounting; Junhao Liu, PhD student, Accounting; Mingyue Zhang, PhD student, Accounting)

Abstract: Interest rates, or broadly speaking, monetary policies, are one important element of macroeconomic conditions and have a substantial influence on microeconomic entities, including companies, financial institutions, and households. In the U.S., the Federal Reserve enacts and communicates monetary policies and interest rate changes through Federal Open Market Committee (FOMC) meetings. As one of the most important information intermediaries in the financial market, financial analysts play an important role in gathering and analyzing information that is relevant for investment, including macroeconomic news from FOMC meetings. Thus, understanding how financial analysts make use of information from FOMC meetings helps investors to better assess the informativeness and quality of financial analyst forecasts. In this study, we employ a set of advanced data analytic techniques to investigate how the interest rate news and the contents of FOMC meetings affect the timing and accuracy of financial analyst forecasts.


“A probabilistic view on understanding time-varying risk premium.”

Peng Liu, PhD Student, Finance (with Bing Han, Professor of Finance, TMX Chair in Capital Markets; Boyu Wang, Assistant Professor, Department of Computer Science, University of Western)

Abstract: The variations in conditional expectations of asset returns (risk premium) are characterized by investors' belief, learning and decision-making processes. The main empirical challenge is that both risk premium and belief cannot be directly observed and measured. Survey data only provides limited and noisy approximations on investor behaviors. The proposed project focuses on empirical aspects of extracting belief and investor behavior patterns from asset returns, which are particularly influential for modelling the dynamics of risk premium as well as for understanding the belief formation process. In my first aim, I use probabilistic machine learning methods to measure the distributions implied by stock market returns. Preliminary results have already shown the risk premium distributions filtered by my framework establish a new channel that connects both predictability and belief dynamics in asset pricing literature. In the second aim, I propose to extend my probabilistic framework to understand the behaviors biases driven by the psychological frictions in investor belief. I hypothesize that developing a framework for understanding the psychological foundations of investor decision-marking process may offer a new channel in both asset pricing and behavioral finance.


“Beyond the Twilight Zone: Restructuring and the Resurrection of Zombie Firms.”

Y. Christine Liu, PhD Student, Accounting

Abstract: Zombie firms are businesses with operating profits that are sustainably insufficient to cover their debt servicing costs. The existence of zombie firms crowds out growth opportunities for other healthy firms and thus distorts overall economic growth. Despite their economic importance, however, the question of what factors enable certain zombie firms to survive while others fail has not yet been addressed in extant literature. I thus propose to focus on an important but hitherto overlooked aspect of zombie firms—their restructuring activities. I first use a financial statement based measure to infer large restructuring charges and find that such accounting charges, that help a firm clean up its business operations and improve transparency, are an important determinant of zombie resurrections. However, charges that reflect earnings management behavior cannot revive zombie firms. I then investigate the types of restructuring activities that determine the revival of zombie firms. I employ a pre-trained deep learning model, Bidirectional Encode Representations from Transformers (FinBERT), to construct measures for business and financial restructuring. I find that business restructuring that improves the operational efficiency of zombie firms and generates revenues helps revive zombie firms, whereas financial restructuring on average does not appear to contribute to their revival. Furthermore, survival analysis suggests that business restructuring shortens the duration of a firm’s zombie status. Finally, I find that recovered zombie firms become more efficient than non-recovered zombie firms. Overall, these results in this study have policy implications for managers, practitioners, and policy makers.

This research has been featured on the Columbia Law School's blog on corporations and the capital markets linked here.


“Strategic Disclosure of Mutual Fund Performance.”

Xijiang Su, PhD Student, Accounting, (with Ole-Kristian Hope, Deloitte Professor of Accounting)

Abstract: Forthcoming.