Events & Seminars

This paper investigates how the presence of the spillover and carryover effects in the multi-channel ad campaign affects the budget allocation decisions of a marketing agency, which strives to maximize the total expected number of clicks or conversions over the campaign. A salient feature of the problem is that the market agency only has access to aggregate data such that the effectiveness of different online advertising channels cannot be estimated using standard methods that typically require individual-level data. The authors propose a data augmentation method for estimating the microlevel consumer advertising response models using aggregate data. The essence of this approach is to simulate latent state dynamics such that the generated data is consistent with the observed aggregate data. The authors then demonstrate the validity of the method using actual channel-level advertising campaign data from an online fashion retailer in Korea. Lastly, the authors study a fluid mean-field formulation and derive key structural insights on the optimal budget allocation policies, which are leveraged to design an implementable budget allocation policy.

Cancer remains one of the leading causes of human death, and early detection is the key to reducing mortality. To detect cancer in the early stages, two-stage screening programs are widely adopted in practice. Individuals receiving positive outcomes in the first-stage (initial) test are recommended to undergo a second-stage test for further diagnosis. The initial test design—i.e., selecting cutoffs to report test outcomes—is crucial for screening effectiveness (i.e., cancer detection) and efficiency (i.e., second-stage capacity costs). However, not all individuals who receive positive outcomes follow up with the second-stage test; evidence shows that adherence behavior is closely associated with the cutoff used in the initial test. This paper studies the initial test design in the context of colorectal cancer (CRC) screening to balance the trade-off between screening effectiveness and efficiency and takes into account individuals’ guideline adherence behavior.

We adopt a Bayesian persuasion framework with information avoidance to model the initial test design and individuals’ response to screening guidelines. We analytically prove that under certain conditions, an initial test using a single cutoff (i.e., a dichotomous test) is optimal for screening follow-up maximization, and a continuous test (i.e., showing exact readings of the biomarker) is optimal for screening effectiveness maximization. We apply the framework to Singapore’s CRC screening guideline design and calibrate the model using various sources of data, including a nationwide survey in Singapore. Our results suggest that compared with the current practice, increasing the cutoff to the level that maximizes expected follow-ups by cancer patients can detect 969 more CRC incidences and prevent 37,820 colonoscopies, which are the second-stage test for CRC screening. Aiming only for high-sensitivity initial tests using lower cutoffs (as in the current practice) can backfire and lead to large numbers of unnecessary colonoscopies and low follow-up rates from cancer patients. We further explore the benefits of using different cutoffs for different subpopulations and use an interpretable clustering technique to construct implementable rules for partitioning the population. We demonstrate that using a lower cutoff for males older than 60 and females older than 70 (high-risk and high-adherence groups) and a higher cutoff for the rest of the screening population (low-risk and low-adherence groups) can further improve screening effectiveness and efficiency.

Cancer remains one of the leading causes of human death, and early detection is the key to reducing mortality. To detect cancer in the early stages, two-stage screening programs are widely adopted in practice. Individuals receiving positive outcomes in the first-stage (initial) test are recommended to undergo a second-stage test for further diagnosis. The initial test design—i.e., selecting cutoffs to report test outcomes—is crucial for screening effectiveness (i.e., cancer detection) and efficiency (i.e., second-stage capacity costs). However, not all individuals who receive positive outcomes follow up with the second-stage test; evidence shows that adherence behavior is closely associated with the cutoff used in the initial test. This paper studies the initial test design in the context of colorectal cancer (CRC) screening to balance the trade-off between screening effectiveness and efficiency and takes into account individuals’ guideline adherence behavior.
We adopt a Bayesian persuasion framework with information avoidance to model the initial test design and individuals’ response to screening guidelines. We analytically prove that under certain conditions, an initial test using a single cutoff (i.e., a dichotomous test) is optimal for screening follow-up maximization, and a continuous test (i.e., showing exact readings of the biomarker) is optimal for screening effectiveness maximization. We apply the framework to Singapore’s CRC screening guideline design and calibrate the model using various sources of data, including a nationwide survey in Singapore. Our results suggest that compared with the current practice, increasing the cutoff to the level that maximizes expected follow-ups by cancer patients can detect 969 more CRC incidences and prevent 37,820 colonoscopies, which are the second-stage test for CRC screening. Aiming only for high-sensitivity initial tests using lower cutoffs (as in the current practice) can backfire and lead to large numbers of unnecessary colonoscopies and low follow-up rates from cancer patients. We further explore the benefits of using different cutoffs for different subpopulations and use an interpretable clustering technique to construct implementable rules for partitioning the population. We demonstrate that using a lower cutoff for males older than 60 and females older than 70 (high-risk and high-adherence groups) and a higher cutoff for the rest of the screening population (low-risk and low-adherence groups) can further improve screening effectiveness and efficiency.

There is an increased interest in bundle selling mechanisms especially with the rise of subscription services. This rise was mainly fueled by the success of subscription services in the digital markets where inventory is unlimited. However, recently there is a slew of subscriptions services that emerged in the retail industry where inventory is limited. In this paper, we take a first step towards understanding the impact of key operational metrics such as inventory levels and limited selling horizons on the optimal bundle selling strategy. We study a dynamic bundle pricing problem when the firm is selling multiple items but with limited inventory. We propose a new scaling regime to study this problem, called high-demand regime, where we scale the arrival rate in order to capture markets where demand is high but inventory is limited. Our results highlight a fundamental limitation of bundling in such markets. Firms should avoid bundling fast moving items together and should rather sell them separately (or bundle fast moving items with slow moving items). Moreover, depending on the tail of the valuation distribution, the firm should either consider static pricing of the items or dynamic pricing. We provide closed form solutions for the static and dynamic pricing policies.

We study the assortment optimization problem and show that local optima are global optima for all discrete choice models that can be represented by the Markov Chain model. We develop a forward greedy heuristic that finds an optimal assortment for the Markov Chain model and runs in $O(n^2)$ iterations. The heuristic has performance bound $1/n$ for any regular choice model which is best possible among polynomial heuristics. We also propose a backward greedy heuristic that is optimal for Markov chain model and requires fewer iterations. Numerical results show that our heuristics performs significantly better than the estimate then optimize method and the revenue-ordered assortment heuristic when the ground truth is a latent class multinomial logit choice model. Based on the greedy heuristics, we develop a learning algorithm that enjoys asymptotic optimal regret for the Markov chain choice model and avoids parameter estimations, focusing instead on binary comparisons of revenues.

Many operations problems are associated with some form of a prediction problem. For instance, one cannot solve a supply chain problem without predicting demand. One cannot solve a shortest path problem without predicting travel times. One cannot solve a personalized pricing problem without predicting consumer valuations. In each of these problems, each instance is characterized by a context (or features). For instance, demand depends on prices and trends, travel times depend on weather and holidays, and consumer valuations depend on user demographics and click history. In this talk, we review recent results on how to solve such contextual optimization problems, with a particular emphasis on techniques that blend the prediction and decision tasks together.

Emission permits are widely adopted to combat climate change and regulatory authorities sometimes allow for the inter-temporal banking and borrowing of emission permits so that firms can flexibly respond to market uncertainties. We find that such time flexibility may lead to poor social performance, especially when the production cost fluctuation is sufficiently large. This result is failed to be captured by the classic simplified assumption where firms cannot sub-exercise emission permits. Furthermore, we demonstrate that the inter-temporal permits transfer should be prohibited when the market is at the red ocean stage or when the pollutant generated relatively instant damage. Lastly, we analyze some restricted permits transfer policies such as transfer discount and transfer cap, which are shown to dominate both the taxation and the non-transferable permits in terms of social welfare.

Information technology (IT) professionals play an important role in firms' IT investments, innovation, and entrepreneurship, contributing to significant economic growth in the U.S. The use of temporary work visas and related immigration policies has attracted a significant controversy and policy debates in the U.S. On the one hand, foreign IT professionals complement domestic IT professionals by facilitating innovation and entrepreneurship. On the other hand, the foreign IT professionals substitute the domestic counterparts by intensifying labor market competition. In this study, we focus on an extension in the Optional Practical Training (OPT) program for STEM graduates from U.S. institutions. Specifically, we explore the effects of the OPT extension on the number and wage of domestic workers in STEM occupations and how these effects differ between IT and non-IT STEM occupations. Our results demonstrate that an increase in the supply of foreign IT professionals from the OPT extension boosts the employment of domestic IT professionals. This study contributes to the information systems, labor economics, and public policy literature by quantifying the impacts of a policy change on the employment of IT professionals and provides rich implications for policymakers.

Even when human point forecasts are less accurate than data-based algorithm predictions, they can still help boost performance by being used as algorithm inputs. Assuming one uses human judgment indirectly in this manner, we propose changing the elicitation question from the traditional direct forecast (DF) to what we call the private information adjustment (PIA): how much the human thinks the algorithm should adjust its forecast to account for information the human has that is unused by the algorithm. Using stylized models with and without random error, we theoretically prove that human random error makes eliciting the PIA lead to more accurate predictions than eliciting the DF. However, this DF-PIA gap does not exist for perfectly consistent forecasters. The DF-PIA gap is increasing in the random error that people make while incorporating public information (data that the algorithm uses) but is decreasing in the random error that people make while incorporating private information (data that only the human can use). In controlled experiments with students and Amazon Mechanical Turk workers, we find support for these hypotheses.

This paper highlights two channels through which blockchain-enabled tokenization can alleviate moral hazard frictions between founders, investors, and users of a platform: token financing and decentralized governance. We consider an entrepreneur who uses outside financing and exerts private effort to build a platform, and users who decide whether to join in response to the platform’s dynamic transaction fee policy. We first show that raising capital by issuing tokens rather than equity mitigates effort under-provision because the payoff to equity investors depends on profit, whereas the payoff to token investors depends on transaction volume, which is less sensitive to effort. Second, we show that decentralized governance associated with tokenization eliminates a potential holdup of platform users, which in turn alleviates the need to provide users with incentives to join, reducing the entrepreneur’s financing burden. The downside of tokenization is that it puts a cap on how much capital the entrepreneur can raise. Namely, if tokens are highly liquid, i.e., they change hands many times per unit of time, their market capitalization is small relative to the NPV of the platform profits, limiting how much money one can raise by issuing tokens rather than equity. If building the platform is expensive, this can distort the capacity investment. The resulting trade-off between the benefits and costs of tokenization leads to several predictions regarding adoption.