Department of Marketing & Entrepreneurship

Time: Friday 10:30 a.m. - 12:00 Noon
Location: 365A Melcher Hall
Open to Public: No reservation or registration required.

Note: Topics and Abstracts will be added to this page throughout the semester

Date Speaker Topic Faculty Host
11/3/2017 Blake McShane
    Multilevel Multivariate Meta-Analysis with Application to Choice Overload
Sam Hui
10/20/2017 Kanishka Misra
UC San Diego
    Dynamic Online Pricing with Incomplete Information Using Multi-Armed Bandit Experiments
  • Click to read Abstract

    Consider the pricing decision for a manager at a large online retailer, that sells millions of products. A manager must decide on real-time prices for each of these products. It is infeasible to have complete knowledge of demand curve for each product. A manager can run price experiments to learn about demand and maximize long run profits. There are two aspects that make this setting different from traditional brick-and-mortar settings. First, due to the number of products the manager must be able to automate pricing. Second, an online retailer can make frequent price changes. In this paper, we propose a dynamic price experimentation policy where the firm has incomplete demand information. For this general setting, we derive a pricing algorithm that balances earning profit immediately and learning for future profits. The proposed approach combines multi-armed bandit (MAB) algorithms statistical machine learning with partial identification of consumer demand from economic theory. Our automated policy solves this problem using a scalable distribution-free algorithm. We show that our method converges to the optimal price faster than standard machine learning MAB solutions to the problem. In a series of Monte Carlo simulations, we show that the proposed approach perform favorably compared to methods in computer science and revenue management.

Seshadri Tirunillai
10/13/2017 Angela Lee
    Goal Focus vs. Temptation Suppression: What Underlies Successful Self-Control?
  • Click to read Abstract

    People facing a self-control dilemma can follow one of three self-control strategies: focus on the goal, counter the temptation, or do both simultaneously. The present research examines the efficacy of these strategies and investigates how people’s regulatory orientation may influence their choice of self-control strategies. Five studies show that (1) focusing on the goal is more effective than countering temptation or the dual strategy of goal focus and temptation countering (2) vigilant goal-focus efforts account for the successful self-control of prevention-oriented individuals and (3) eagerness efforts that underlie the dual self-control strategy ironically weakens the self-control of promotion-oriented individuals.

Melanie Rudd
10/9/2017 Garrett Johnson
    The Online Display Ad Effectiveness Funnel & Carry-Over: A Meta-study of Predicted Ghost Ad Experiments
Seshadri Tirunillai
10/2/2017 Abhinav Uppal
    A Theory of Selling Formats in Retailing: Direct versus Mediated Access
  • Click to read Abstract

    Retailers worldwide employ various selling formats characterized by different degrees to which customers can access and inspect products in the store. In the ``direct access'''' format, all available products are stocked on shelves directly accessible to customers for inspection, while store associates offer minimal assistance. In the ``mediated access'''' format, retail stores are manned by shopkeepers who offer one product at a time to customers and the customers decide whether to purchase an offered product or to ask for an alternative. We build a theoretical model in which a retailer makes selling format, product assortment and pricing decisions, and consumers have shopping costs. There are two products: a general purpose brand that provides the same utility uniformly to all consumers, and a specialized brand that gives ex ante uncertain utility to a consumer that can be higher or lower than the utility of the general purpose brand, and a consumer can resolve this by inspecting this brand. We find that the retailer chooses the mediated access selling format with the specialized brand offered first when customers'' uncertainty about fit with the specialized brand is large (as long as the retailer''s margin on the general purpose brand is not too high). If consumers'' uncertainty about fit with the specialized brand is medium, the retailer chooses to internalize consumer shopping costs by employing the direct access format and carries both brands. If consumers'' uncertainty about fit with the specialized brand is small, the retailer chooses the mediated access format carrying only the general purpose brand. Our model offers an explanation for the observation that the mediated access selling format is more popular in emerging markets (as compared to developed markets) where consumers'' shopping costs (e.g., cost of time) are typically small, but in these markets this format is less popular for large, organized retailers (as compared to small, unorganized retailers) that may be able to obtain better trading terms, e.g., larger retail margins, from upstream sellers.

Seshadri Tirunillai
9/29/2017 David Schweidel
    How Content Affects Clicks: A Dynamic Model of Online Content Consumption
  • Click to read Abstract

    With many consumers being exposed to news via social media platforms, news organizations are challenged to attract visitors and generate revenue during visits to their websites. For advertising-supported websites, website traffic is related directly to revenue. Using clickstream data from a major news organization, the authors develop a user-level dynamic model of clickstream behavior that takes into account the content of both headlines and stories that visitors read. The authors assume that article headlines inform readers’ expectations of the article content and that prior consumption of articles during a visit to the website will affect preferences for subsequent consumption. The authors find that readers appear to exhibit state dependence in the tone of the articles that they read. They also show how the topics expressed in headlines can affect the amount of content readers consume when visiting the news organization to a much larger degree compared to the topics expressed in the content of the article. Online publishers can make use of such findings to present visitors with content that is likely to maintain and/or increase their engagement and consequently drive advertising revenue.

Sam Hui
CBB 524
Shijie Lu
    Dynamic Effect of Digital Badging on User-generated Content Platforms
  • Click to read Abstract

    User-generated content (UGC) platforms have increasingly utilized digital badging as a key mechanism to incentivize user engagement. This research empirically examines the dynamic effect of digital badging on individual user’s content consumption, interaction, and generation behavior on a UGC platform. Two types of digital badges are considered: digital badges awarded by the platform (FDBs) and digital badges received from peer users (UDBs). To capture the dynamics in user engagement, the authors develop a hidden Markov model in which a user’s engagement decisions are affected by her underlying engagement state associated with the platform and digital badges received. By applying the proposed model to a unique panel data set from a large UGC platform, the authors find three distinctive engagement states of users. In the short run, while the impact of UDBs is generally homogenous by encouraging both content consumption and generation for users in all three states, the impact of FDBs is more heterogeneous across users’ engagement states. In the long run, a double-edged effect of FDBs is found by encouraging content generation but suppressing content consumption. Finally, the authors demonstrate the value of the proposed model in helping UGC platforms design targeted FDB systems.

Seshadri Tirunillai
9/22/2017 Kathleen (Kathy) Li
    Statistical Inference for Average Treatment Effects Estimated by Synthetic Control Methods
  • Click to read Abstract

    The synthetic control method (SCM), a powerful tool for estimating average treatment effects (ATE), is increasingly popular in fields such as statistics, economics, and marketing and has been called “arguably the most important innovation in the evaluation literature in the last fifteen years” (Athey and Imbens 2016). However, SCM has the main limitation that there is no inference theory: therefore, it is not possible to calculate confidence bounds, exact p-values or conduct hypothesis tests. Existing work mostly uses placebo tests. I derive the inference theory for the synthetic control ATE estimators using projection theory, and show that a properly designed subsampling method can be used to obtain confidence intervals and conduct hypothesis test, whereas the standard bootstrap cannot. A second limitation of SCM is that when there is heterogeneity among control units and treatment unit (which occurs frequently in economic and marketing data settings), the synthetic control version of parallel lines assumption may not hold and it can perform poorly in in-sample fit and out-of-sample predictions. I show through an empirical illustration that a modification to the SCM proposed by Doudchenko and Imbens (2016) allows the method to be more widely applied. Simulations and an empirical application examining the effect of opening a physical showroom by Warby Parker demonstrates the usefulness of the inference theory and modified synthetic control method.

Seshadri Tirunillai
365B MH
Tong (Joy) Lu
    Binge Consumption of Online Content: A Boundedly Rational Model of Goal Progress and Knowledge Accumulation
  • Click to read Abstract

    Binge consumption of online content has emerged as a trending phenomenon among customers of online streaming services, with various content providers spanning the spectrum from entertainment to education. Here, we focus on binging within an online education setting, using clickstream data from Coursera in which we observe individual-level lecture and quiz consumption patterns across multiple courses. We extend the literature by distinguishing between “temporal binging,” where individuals consume multiple pieces of content in a single sitting, and “content binging,” where individuals consume content from the same course in succession. We build a model that captures individual decisions about which course to consume, whether the content is a lecture or a quiz, and when to take breaks of different lengths. The parameters of our model can be mapped to specific theories in consumer psychology, which allows us to test for the mechanisms that drive binge consumption. There are three key features of our model: First, we assume that individuals are motivated to consume to progress towards the goal of completing the course. We find that consumption patterns are consistent with those predicted by goal gradient and slack theory, but inconsistent with other theories. Second, individuals are motivated to accumulate knowledge through lectures in order to pass quizzes. Beliefs about quiz-taking abilities are updated based on prior quiz scores, which we observe as a concrete measure of knowledge of the course material. Third, we assume that individuals are forward-looking in a boundedly rational way. To explore whether different firm policies may impact binging patterns, we conduct counterfactual simulations to determine how the timing of content release affects consumption and knowledge accumulation. We then test these predictions using data obtained after a “natural experiment” policy change when the Coursera platform transitioned from sequential to simultaneous course content release in the year following our original data sample.

Seshadri Tirunillai
7/31/2017 Julie Lee
University of Western Australia
    The theory, measurement and application of values and values-expression across the human lifespan
  • Click to read Abstract

    Personal values, as a major aspect of personality, have received a great deal of attention in psychology, but far less in marketing. Values, as motivational life goals, indicate what is important to people across contexts and situations. They have been related to a wide range of attitudes and behaviours, especially those that are not resource constrained. However, marketers are mostly interested in behaviour that is resource constrained. In this seminar, recent refinements to values theory and measurement will be discussed and the ways in which value-expression might be uncovered will be explored. Data presented is from the first stage of a large cross-sequential study of over 7000 adults, who are participating in a 3 year study. The full study will be explained and opportunities for collaboration can be explored.

Jackie Kacen
Seminar Series
BBA in Marketing
BBA in Entrepreneurship
Program for Excellence in Selling
Master of Science in Marketing
Ph.D. in Marketing
Course Descriptions
News & Awards
Centers & Institutes
Staff Directory
Contact Us

Department of Marketing & Entrepreneurship
University of Houston
334 Melcher Hall
Houston, Texas 77204-6021
Phone: 713-743-4555
Fax: 713-743-4572