Category: Academic Departments
From “A Hidden Markov Model for Collaborative Filtering,” MIS Quarterly, 36(4), 1329-1356.
Commercial websites are constantly suggesting new products and content to us—a mechanized, cyber-age form of the old urging, “if you liked that, you’ll love this!” In tech terms, the systems that generate these suggestions are called personalized recommender systems. But how can these computer systems account for the age-old human tendency to change our desires as time goes on?
A new study by Boston University’s Nachiketa Sahoo and co-authors Param Vir Singh and Tridas Mukhopadhyay is one of the first to address this problem.
Sahoo is an assistant professor in information systems at Boston University School of Management; Singh and Mukhopadhyay are faculty members at the David A. Tepper School of Business at Carnegie Mellon University. Their paper, “A Hidden Markov Model for Collaborative Filtering,” appearing in MIS Quarterly‘s special issue on business intelligence research, suggests using a stochastic algorithm called a hidden Markov model (HMM) to process data about user activity and preferences, rather than the common algorithms used now by most personalized recommender systems. The authors show that the HMM, a more dynamic model, allows online personalized recommender systems to account for changing user preferences.
A New Model to Address Changing User Preferences
The authors point out that dynamic, not static, user tastes and desires are integral to the consumer experience, particularly with the repeat consumption of so-called “experience goods,” such as movies, music, and news. “This causes problems for a recommender system that has been trained to identify customers’ preferences from their past ratings of products,” the authors write.
Sahoo et al. propose a customized HMM algorithm to estimate user preferences and make recommendations. They evaluate their approaches using three real-world datasets: one containing employees’ blog reading activity in a Fortune 500 IT services firm, one documenting users’ movie watching behavior in the Netflix Prize dataset, and one tracking users’ music listening behavior on last.fm. Comparing the performance of their algorithm with that of several other popular algorithms in recommender systems, the authors show that the HMM-based algorithm performs as well or better than the other algorithms, particularly as user preferences change.
Their approach is based on the intuition that older data, rather than being discounted—as they are in some current personalized recommender systems—could instead be used to learn about that user’s preference and then applied to another user. “Data from a user’s past may not be useful for making recommendation for the user now,” they argue, since “her preference has changed, but it might be useful for making a recommendation for someone who currently has that preference.”
Read more about ”A Hidden Markov Model for Collaborative Filtering.”
Banner photo is a visualization of related movies found by a computer algorithm created for Netflix Prize. Each movie is represented by a dot, and colored lines signify a similarity between pairs. Photo courtesy of flickr user chef_ele.
On January 10, Jacqueline J. and Arthur S. Bahr Professor of Management Michael Salinger‘s piece “Why the FTC Was Right Not to Sue Google” was featured in the Forbes Leadership Forum on Forbes.com. Salinger, a professor in the Markets, Public Policy & Law department, is a former Director of the Bureau of Economics at the United States Federal Trade Commission.
January 3 should go down as one of the most important and proudest moments in the history of United States antitrust enforcement. After a 19-month inquiry, the Federal Trade Commission announced that it had voted unanimously to close its investigation into the design of Google’s search results. The FTC’s decision is a victory for Google, a defeat for those who tried to persuade the FTC to use the antitrust laws to hinder rather than promote competition, and a victory for Google users. It is not easy for a law enforcement agency to devote substantial resources to an investigation and then not bring a case, but sound antitrust enforcement dictates that it must do so when, as happened here, the investigation failed to uncover evidence of a violation.
To understand what was at stake in the case, go to Google and enter a query for “New York weather.” The top result will say “Weather for New York, NY,” with a minimal four-day forecast that may be sufficient for some users. Just below that will be links to sites that provide more detailed weather information. To the extent that users find the information provided directly by Google to be sufficient, weather sites might get less traffic. But Google users are better off, and that is the key point. As FTC Chairman Jon Leibowitz explained about the FTC’s decision, the antitrust laws are supposed to protect competition, not individual competitors. And, far from being an antitrust violation, improving search results to get users the information they need is precisely the sort of competition the antitrust laws are supposed to encourage.
Read Salinger’s full piece on Forbes.com.
Banner image courtesy of flickr user Robert Scoble.
Karim, Samina (2012). “Exploring structural embeddedness of product market activities and resources within business units.“ Strategic Organization 10(4): 333-365.
The lead article in the November 2012 issue of Strategic Organization is Samina Karim‘s study “Exploring structural embeddedness of product market activities and resources within business units.”
Karim is an assistant professor in strategy and innovation at Boston University School of Management.
This paper defines “embeddedness” as the dependence on routines and coordination mechanisms within one’s own business unit, and explores the degree to which embeddedness impacts the success of product market activities (PMA). Karim focuses on which alternative better supports the longevity of a product market within the firm: 1) moving a PMA out of one unit and into another, or 2) moving the entire unit with its PMA into another unit (so that the PMA is still managed in its original organizational context, even though it is now “housed” in another, bigger unit).
Among Karim’s findings:
- If activities and resources are highly embedded in their business units, then “reconfiguring” a unit (by adding to it, trimming it down, or recombining it with another unit) may have consequences on how successful the firm is at its product market activities.
- Moving an entire unit with its PMA into another unit leads to a greater likelihood of retaining the PMA (i.e. the PMA will not be not divested or shut down).
For managers within PMA units, the practical insights of Karim’s study include:
- If managers are going to move a PMA from one unit to another, they are better off recombining the entire unit into the other than simply moving the PMA from one to the other.
- If managers are going to move a PMA from one unit to another, they should be less worried about whether the PMA was acquired or not, and more focused on whether, during the move, they can keep intact the PMA’s former unit’s routines and processes.
“Pricing Regulation and Imperfect Competition on the Massachusetts Health Insurance Exchange”
A new National Bureau of Economic Research (NBER) study, authored by Keith M. Marzilli Ericson and Amanda Starc and focused on pricing regulation in health insurance exchanges (HIE), shows that purchasing mandates can be essential to the functioning of this entire market.
Ericson is an assistant professor of markets, public policy, and law at Boston University School of Management. Starc is an assistant professor of health care management at the Wharton School at the University of Pennsylvania.
Their study, “Pricing Regulation and Imperfect Competition on the Massachusetts Health Insurance Exchange,” explores pricing regulation, consumer demand, and insurer profits in HIE, which are government-run marketplaces for private insurance. The authors use data from Massachusetts’s HIE, the first in the nation, and then apply these data to the broader functioning of health exchanges themselves. Their focus on the mandate, requiring citizens to purchase a minimum level of insurance, sheds light on one of the most controversial issues in Congress’ recent struggles over health care across America.
HIEs: An Ideal Context for Exploring Consumer Welfare, Regulation, and Profit
The authors point out that HIEs offer an ideal opportunity to study issues of consumer welfare, competition, government regulation, and firm profits, as they offer a wide range of choice to consumers in the context of a heavily regulated environment. Moreover, in the next few years, a projected 20 million Americans across the country will purchase health insurance through these exchanges, as the 2010 Affordable Care Act has mandated that states and the federal government develop HIEs.
But Ericson and Starc note a lack of previous research exploring how insurance pricing regulation actually functions in markets where firms have some market power to charge prices above their costs—a condition they refer to as “imperfect competition.”
Their new NBER study fills this gap.
“If the Mandate Is Removed, Markets Can Unravel”
Ericson and Starc first execute a series of simulations based on data from the Massachusetts HIE to show how changing regulations on insurers can vary prices between different types of consumers (such as older vs. younger consumers) and can impact other important and controversial insurance market regulations, such as minimum loss ratios (which attempt to limit insurer profits), risk adjustment (which attempts to equalize insurers’ costs derived from insuring different populations), and mandated insurance purchase (which attempts to ensure market participation).
Ultimately, the study’s simulations show that if the mandate is removed, markets can unravel, due to differences in preferences across a broad population where a significant segment of that population would be willing to withdraw from the market altogether if they can’t find a price they are willing to pay.
If consumers are allowed to opt out of coverage, the authors note, the most price-sensitive consumers—who tend to be both young and relatively healthy—will tend to opt out. As these consumers opt out, the less price-sensitive consumers—who tend to be both older and have higher health costs—are the ones remaining in the market, which in turn leads to higher markups. If enough people are willing to drop out of the market altogether, the authors note, “a death spiral” can occur. “As a result,” Ericson and Starc show, “a weak or absent mandate may negate the consumer surplus gains achieved” from the other regulations still in place.
Read more about the study “Pricing Regulation and Imperfect Competition on the Massachusetts Health Insurance Exchange.”
Banner photo courtesy of flickr user Images_of_Money.
From MIT Sloan Management Review‘s Innovation Issue
In the Winter 2013 “Innovation” issue of MIT Sloan Management Review, Senior Associate Dean, Professor in Organizational Behavior, and Everett W. Lord Distinguished Faculty Scholar Karen Golden-Biddle explores a new approach to organizational transformation and meaningful change:
Too often, conventional approaches to organizational transformation resemble the Big Bang theory. Change occurs all at once, on a large scale and often in response to crisis. These approaches assume that people need to be jolted out of complacency to embrace new ideas and practices. To make that happen, senior management creates a sense of urgency or takes dramatic action to trigger change. Frequently, the jolt comes from a new CEO eager to put his or her stamp on the organization. Yet we know from a great deal of experience that Big Bang transformation attempts often fail, fostering employee discontent and producing mediocre solutions with little lasting impact.
But meaningful change need not happen this way. Instead of undertaking a risky, large-scale makeover, organizations can seed transformation by collectively uncovering “everyday disconnects” — the disparities between our expectations about how work is carried out and how it actually is. The discovery of such disconnects encourages people to think about how the work might be done differently. Continuously pursuing these smaller-scale changes — and then weaving them together — offers a practical middle path between large-scale transformation and small-scale pilot projects that run the risk of producing too little too late.
Researchers tend to overlook this option because few managers have employed it until recently, assuming they needed to take an all (Big Bang) or small (pilot projects sequestered away from the dominant organizational culture) approach to organization change. That may have been more true in the past when organization boundaries were less malleable, communication more difficult and people less mobile. However, today’s complex and connected global environment makes step-by-step transformation by managers inside most organizations a real possibility…
Banner photo courtesy of flickr user Yogesh Mhatre.
New Study Uncovers Green Eco-Seals’ Opposing Impact on Different Consumer Types
Researchers Barbara Bickart and Julie Ruth have completed a study filling a crucial gap in advertisers’ knowledge about the efficacy of green marketing techniques such as eco-seals, showing that they have a distinctly different impact—and in fact sometimes opposing effects—on different types of consumers.
Bickart and Ruth are associate professors in marketing at Boston University School of Management and Rutgers University, respectively. Their new study, “Green Eco-seals and Advertising Persuasion,” is forthcoming in the Journal of Advertising‘s special issue on green advertising.
Bickart and Ruth focus on the differing persuasiveness of eco-seals for consumers with high versus low concern about environmental issues, as well as with high versus low familiarity with a brand. They also offer insight into how these different consumers react depending on an eco-seal’s source and the type of specific messaging it provides.
Among their findings:
- When consumers have a low-level of environmental concern, the presence or absence of an eco-seal on a package has limited impact on purchase intentions, regardless of familiarity with the brand, although;
- When consumers have a low-level of environmental concern, the absence of a seal leads them to evaluate the familiar brand more favorably than the unfamiliar brand.
- When a consumer has a high-level of environmental concern, eco-seals in general generate more favorable purchase intentions for familiar brands, although eco-seals with an ambiguous source generate less favorable purchase intentions for unfamiliar brands, and perhaps most surprising;
- High-concern consumers are more likely to respond favorably to eco-seals generated by the manufacturer, as opposed to an independent source such as the government, suggesting that familiar-brand seals boost these consumers’ beliefs about a company’s concern for the environment.
As a whole, the study data points to numerous specific strategies for marketers and policy makers about the most effective use of eco-seals and message strategies for various easily-identifiable target audiences.
See a recent profile of this research at the Wall Street Journal blog, “Corporate Intelligence.”
Banner photo courtesy of flickr user Pylon757.
In their recent article “Financial Literacy 101,” offering experts’ top recommendations for novice investors, The Wall Street Journal spotlighted Risk Less and Prosper by Boston University’s Zvi Bodie, the Norman and Adele Barron Professor of Management, and co-author Rachelle Taqqu:
With its focus on goal-based investing, this book offers concrete steps to help beginning investors detail their specific needs and wants for the future, and to invest based on those goals.
Zvi Bodie, a management professor at Boston University, advises investors to take on risk only with money they can afford to lose. For the rest, he recommends specific inflation-indexed government bonds.
“Stocks can be a winning strategy, but they can also bring tragedy, and Bodie carefully sets out the risks and rewards of the alternatives,” says Dallas Salisbury, chief executive of the Employee Benefit Research Institute, a nonprofit think tank.
HBS Working Knowledge Article Spotlights “Public Procurement and the Private Supply of Green Buildings”
Timothy Simcoe and Michael W. Toffel have published a new study on how government policies can stimulate private demand for environmentally friendly buildings. Simcoe is an assistant professor in the Strategy and Innovation Department at Boston University School of Management. Toffel is an associate professor in the Technology and Operations Management group at Harvard Business School.
Their study, “Public Procurement and the Private Supply of Green Buildings” has been published by the National Bureau of Economic Research (NBER Working Paper Number 18385) and was recently spotlighted in the article “LEED-ing by Example,” from HBS Working Knowledge:
In the debate over whether to increase or decrease the stringency of environmental regulations, the possibility that government agencies might use purchasing to stimulate market demand for “green” products and services is often overlooked. Nevertheless, several recent US presidents (of both parties) have issued executive orders requiring federal agencies to use environmentally preferable products and services whenever possible…
But….there has been virtually no industry analysis of whether this strategy actually worked, up until now.
In a new paper, “Public Procurement and the Private Supply of Green Buildings,” authors Timothy Simcoe and Michael W. Toffel show that there is, indeed, a spillover effect to the private sector. The authors studied what happened after municipal governments in California adopted policies that required public (but not private) building renovations and new construction to build “green,” which nearly always meant adhering to the US Green Building Council’s Leadership in Energy and Environmental Design (LEED) standard. After local governments decided to pursue LEED certification for their own buildings, there was an uptick in the number of local architects, general contractors, and other construction industry professionals who sought LEED accreditation. Also, the use of the LEED standard increased among private builders in the same local markets.
Read more at Working Knowledge‘s LEED-ing by Example.
See a recent overview of this research on Forbes.com
Above: Photo of the first LEED-certified parking structure in the US by flickr user Schlüsselbein2007.
Placed 16th overall, 1st among female academics
In December 2012, the American Marketing Association (AMA) launched a new annual initiative to track top contributors to premier marketing journals such as the Journal of Consumer Research, Journal of Marketing, Journal of Marketing Research, and Marketing Science. The goal is to acknowledge the most productive researchers, both by authorship and university affiliation, in the previous five years and to provide a unique perspective to future doctoral students making application decisions about marketing PhD programs.
In the first Author Productivity in the Premier AMA Journals list, Boston University School of Management’s Shuba Srinivasan has ranked number 16 overall and number one for female academics for contributions to the Journal of Marketing and Journal of Marketing Research.
Srinivasan is an associate professor, Dean’s Research Fellow, and PhD Program faculty liaison in marketing. Her research focuses on strategic marketing problems, the link between marketing and financial gains, and metrics for gauging marketing performance.
Professor Srinivasan’s publications include:
- Srinivasan, S., K. Pauwels, and V. Nijs (2008), “Demand-based Pricing Versus Past-price Dependence: A Cost-Benefit Analysis,” Journal of Marketing, 72 (2), 15-27. (Abstract)
- Srinivasan, S. and D. M. Hanssens (2009), “Marketing and Firm Value: Metrics, Methods, Findings and Future Directions,” Journal of Marketing Research, 46 (3), 293-312. (Abstract)
- Srinivasan, S., K. Pauwels, J. Silva-Risso, and D. M. Hanssens (2009), “Product Innovation, Advertising Spending and Stock Market Returns,” Journal of Marketing, 73 (1), 24-43. (Abstract)
- Srinivasan S., M. Vanhuele, and K. Pauwels (2010), “Mind-Set Metrics in Market Response Models: An Integrative Approach,” Journal of Marketing Research, 47 (4), 672-684. (Abstract)
- Osinga, E., P. Leeflang, S. Srinivasan, and J. Wierenga (2011), “Why Do Firms Invest in Consumer Advertising with Limited Sales Response?” Journal of Marketing, 75 (1), 109–124. (Abstract)
Professor Srinivasan is also chair of the AMA’s Marketing Research Special Interest Group and serves on the editorial boards of Marketing Science, Journal of Marketing Research, and International Journal of Research in Marketing. Among her other honors are being named a finalist for the 2012 Robert D. Buzzell Best Paper Award, winning the 2010 Broderick Prize for excellence in research scholarship at Boston University’s School of Management, and receiving the 2001 European Marketing Academy Best Paper Award.
See Professor Srinivasan’s additional honors.
Associate Professor and Dean’s Research Fellow gives advice in BU Today about the most common shopper mistakes and how to avoid them
Excerpts from BU Today:
The future of the economy may be uncertain, but money woes appear not to have dampened the spirits of holiday shoppers. Spending over the four-day weekend following Thanksgiving, including Black Friday and Cyber Monday, reached $59 billion, a 13 percent increase over last year, according to the National Retail Federation. The organization predicts that holiday sales will jump 4 percent over last year’s number, to $586 billion.
What does it all mean? Why do people spend more when they may have less? How can shoppers get the biggest bang for their buck? BU Today spoke with Barbara Bickart, a School of Management associate professor of marketing and a Dean’s Research Fellow, about common holiday shopping pitfalls, why we spend irrationally this time of year, whether we should feel guilty about buying for ourselves, and why we should buy the same gift for everyone on our holiday list.
What are some common mistakes shoppers make during the holidays?
The real issue is being tempted by deals that are right in front of you that seem too good to resist and look like they’re going to go away tomorrow. The retailers do a really good job of trying to convey that this is a limited time offer, that it’s a very precious, valuable, scarce deal, and that there are only so many of these available. So people think, I’ve got to act now.
How can we avoid these pitfalls?
One thing is having a list and knowing exactly what you’re going to get. And if you’re going to get things for yourself, know what those are too, because you’re probably going to be more impulsive for yourself than you are for others.
Another thing is not to go shopping when you’re tired or depleted, because as we make many decisions, we start to become more depleted and then we become more inclined to be impulsive. If you go shopping at a time when it’s not so busy or when you can be energetic, that’s going to help you avoid making those impulsive decisions. A shopping marathon is not a good idea. Do a little bit at a time and maybe do it on a Tuesday night when the stores aren’t so crowded.
See all of Associate Professor Bickart’s tips on BU Today.
Photo by Flickr user ThomasOfNorway.