Scholar’s Applied Research Projects
Because analytics does not belong to one discipline, we are moving beyond the traditional sectors and reaching out to researchers, graduate students and practitioners whose interests cross disciplines. We have assembled a team of researchers from Management Sciences, Operations Management, Marketing, Information Systems, Finance, Business Economics and Computing Science. These diverse backgrounds will provide a variety of perspectives in our analytics research. This diversity will add to the creativity of our research and our results.
The Impact of Inventory Risk on Market Prices Under Competition
By A. Ovchinnikov
This paper shows that due to inventory risk, an increase in the number of competitors can lead to an increasing trend in market prices. Furthermore, because of how inventory risk impacts competitive behavior, firms may prefer to incur inventory risk rather than to avoid it.
Anti-discrimination Laws, AI, and Gender Bias: A Case Study in Non-mortgage Fintech Lending
By A. Ovchinnikov
This paper extends the conceptual understanding of model-based discrimination from computer science to a realistic context that simulates the situations faced by fintech lenders in practice, where advanced machine learning (ML) techniques are used with high-dimensional, feature rich, highly multi-collinear data.
Why You Should Allow Returns on Customized Products
By A. Ovchinnikov
Consumers are accustomed to generous product returns policies online. Firms recognize that it makes the purchase decision easier. But most companies restrict these policies to standard products and customized products are typically not returnable. In this article the authors argue that this is a mistake.
What Happens When AI is Used to Set Grades
By A. Ovchinnikov
In 2020, with high school exams canceled in many countries, the International Baccalaureate Organization (IBO) deployed an AI to determine final grades based on current and historical data. This article highlights the risks of delegating life-altering decisions to AI without considering how apparently anomalous decisions can be appealed and, if necessary, changed.
Improving Fairness in Credit Lending Models using Subgroup Threshold Optimization
By S. Thomas
This paper introduces a new fairness technique, called Subgroup Threshold Optimizer (STO). STO works by optimizing the classification thresholds for individual subgroups in order to minimize the overall discrimination score between them. Our experiments on a real-world credit lending dataset show that STO can reduce gender discrimination by over 90%.
Joint Product Framing and Order Fulfillment under the MNL Model for E-Commerce Retailers
By Y. Lei
Using data from a major U.S. retailer, this paper shows that jointly planning product framing and order fulfillment can have a significant impact on online retailers’ profitability. This is a technically challenging problem and this paper makes progress toward resolving this challenge using techniques such as randomized algorithms and graph-based algorithms.
Expected Profit of Fixed Price Policy Decays Exponentially in the Lead Time
By Y. Lei
While it is intuitive that dynamic pricing can yield a better expected profit than the best Fixed Price (FP) policy because the former is more flexible than the latter, the magnitude of this improvement is not well understood. In this paper, we shed partial light on this question by focusing on the impact of delivery lead time in a setting with a linear purchase rate and a Normal-like demand.
The impact of discriminatory pricing based on customer risk
By M. Nediak and C. Kolsarici
Using individual-level loan application and approval data and segment-level customer risk as the price discrimination criterion for a firm, the authors develop a three-stage model that accounts for the salesperson’s price decision. The authors compare the profitability of this sales force price delegation model to that of a segment-level centralized pricing model where agent incentives and consumer prices are simultaneously optimized.
The Anatomy of the Advertising Budget Decision
By C. Kolsarici
Using analytical approach, this paper investigates how advertising budget is spent by looking at 4 components: (1) baseline spending, (2) adaptive experimentation, (3) advertising-to-sales ratio, and (4) competitive parity. The paper proposes a methodology to estimate and infer the weights of these four components. The adaptive experimentation finding, combined with evidence on the use of heuristic methods, suggests that budget decision making is characterized by bounded rationality.
Selling Passes to Strategic Customers
By J. Wang, M. Nediak and Y. Levin
Under the control-theoretic framework, authors find that the optimal pricing policy has a turnpike property; the optimal price trajectories stay near the steady state for most of the sales horizon, and the fixed-pricing policy performs very well when the horizon is long enough. In the turnpike, we show that passes should offer a quantity discount when customers are not fully strategic. As a form of advance purchase, passes allow the firm to capitalize on strategic behavior without limiting the supply.
Are You Ready for a ChatGPT World?
S. Thomas
Since it launched last November, ChatGPT has become the talk of the town. It can write articles, messages, essays and songs, using machine-learning algorithms. But how good is ChatGPT? What is it lousy at? And how will such high-powered AI applications fit into everyday business operations?
When Analytics Gets Personal
Y. Lei
Prescriptive analytics has important applications across a variety of sectors — from helping companies develop and improve their products to helping banks identify fraudulent purchases. But this form of business analytics also raises challenging privacy issues. This article explains how firms can better approach privacy risks while still making data-driven personalized decisions.
Data Analysis Using Spark and Sparkling Water
By J. Zhang
This paper introduces Sparkling Water which is an integration of a general purposed big data framework called Spark, and a powerful machine learning library called H2O.ai.
Strategic Analytics
By Yuri Levin, Jeffrey Mcgill
With the recent growth in availability of data and inexpensive analytical tools, forward-thinking enterprises can mine the potential of these powerful methods for themselves. Here is what you need to know to get started.
Revenue Management, Then And Now
Interview by Alan Morantz
From the basement to the boardroom: The analytics revolution is returning us to the Marrakech markets of old.
Research Brief: Do Multi-Media Advertising Campaigns Pay Off?
By Ceren Kolsarici, Demetrios Vakratsas
In the era of Big Data, marketing managers must move beyond analyzing aggregates of market data.
Name-Your-Own-Price Sales Channels: Revenge Of The Hive
By Alan Morantz
The Pricelines and Hotwires of the world are vulnerable to online consumers gaming the system by banding together. Fortunately for the companies, consumers are loners.
Weighing Big Pharma’s Consumer Pitch
By Anna Sharratt
Direct-to-consumer advertising gets a bad rap in Canada. But is it really such a negative force?
Out-Stocking Your Competitor
By Alan Morantz
When it comes to products with a short shelf life, replenishing your inventory by instinct alone makes you easy pickings. Time to build your analytics toolkit.
Dynamic Story: Torquing The Hollywood Promo Machine
By Alan Morantz
When a new movie or book is rolled out in different formats over time, how do firms get the biggest bang from their advertising and viral campaigns?
Markdowns: Decoding The Nuances Of Consumer Behaviour
By Alan Morantz
A new model that accounts for how consumers view wait-or-buy decisions shows retailers how to harvest higher revenue from deeper discounts.