- Consumer Market Behavior and Pricing
- Digital Platforms and Economics
- Digital Marketing and Social Media
- Advanced Causal Inference Techniques
- Transportation and Mobility Innovations
- Innovation Diffusion and Forecasting
- Insurance and Financial Risk Management
- Sharing Economy and Platforms
- Wine Industry and Tourism
- Business Strategy and Innovation
- Auction Theory and Applications
- Consumer Behavior in Brand Consumption and Identification
- Data Stream Mining Techniques
- Transportation Planning and Optimization
- Consumer Retail Behavior Studies
- Advanced Bandit Algorithms Research
- Recommender Systems and Techniques
- Intellectual Property and Patents
- Cardiac Fibrosis and Remodeling
- Spam and Phishing Detection
- Survey Methodology and Nonresponse
- Qualitative Comparative Analysis Research
- Pharmaceutical Economics and Policy
- Game Theory and Applications
- Forecasting Techniques and Applications
University College London
2017-2024
Peking University
2008-2022
Johns Hopkins University
2022
École Supérieure des Sciences Économiques et Commerciales
2022
University of Hong Kong
2022
Cornell University
2022
University of Southern California
2022
Western University
2022
University of Colorado Boulder
2022
Western University of Health Sciences
2022
Television, the predominant advertising medium, is being transformed by microtargeting capabilities of set-top boxes (STBs). By procuring impressions at STB level (often denoted “programmatic television”), advertisers can now lower per-exposure costs and/or reach viewers most responsive to creatives. Accordingly, this study uses a proprietary, household-level, single-source data set develop an instantaneous show and advertisement viewing model forecast consumers' exposure downstream...
“Freemium,” whereby a basic service level is provided free of charge but consumers are charged for more advanced features, has become popular business model firms selling digital goods. However, it not clear whether the launch version helps or hurts demand an existing paid version. The may allow to sample product before making purchase decision and subsequently increase version, also cannibalize We use comprehensive data set on game apps from Apple’s App Store that tracks both versions...
Panel studies typically suffer from attrition, which reduces sample size and can result in biased inferences. It is impossible to know whether or not the attrition causes bias observed panel data alone. Refreshment samples—new, randomly sampled respondents given questionnaire at same time as a subsequent wave of panel—offer information that be used diagnose adjust for due attrition. We review bolster case use refreshment samples studies. include examples both fully Bayesian approach...
E-commerce platforms have an informational advantage over their third-party sellers, leading to the common belief that a platform's market entry would harm sellers with similar products. However, unlike traditional retail competition, platform and its aligned incentives: commission depends on seller's revenue, rely strengthen online presence. Hence, has no incentive enter revenue severely. This paper introduces duopoly model where seller is "incumbent" "potential entrant". Our captures two...
This paper considers the history of keywords used in Marketing Science to develop insights on evolution marketing science. Several findings emerge. First, “pricing” and “game theory” are most ubiquitous words. More generally, three C's four P's predominate, suggesting that common practical frameworks align. Various trends exist. Some words, like “pricing,” remain popular over time. Others, “hierarchical Bayes,” have become more popular. Finally, some words superseded by others, “diffusion”...
Because of the informational advantage online marketplaces (i.e., platforms), it is a common belief that platform’s market entry will be detrimental to third-party sellers who sell similar products on platform. To examine validity this belief, we conduct an exploratory analysis using sales data for single product category provided by JD.com month March 2018. Our reveals unexpected result. Upon entry, can afford charge higher price, obtain demand, and earn profit. provide plausible...
Abstract Surge pricing has been used to coordinate supply and demand in the ride‐sharing industry, but its causal effects on driver behavior remain unclear. This motivates us examine how surge causally affects earnings labor by leveraging a unique quasi‐experiment, which leading company China introduced two cities at different times. Using difference‐in‐differences design with forest method, we find that led increases drivers' weekly revenue. Decomposing revenue into “intensive margin”...
Expert opinion exerts tremendous influence on the purchase journey, but its effect overall consumer experience is ambiguous as it can give rise to both "expectation" and "reputation" effects. This paper explores of expert opinions via lens reviews in restaurant industry, where are conveyed by Michelin stars. We construct a unique data set based Guide for Great Britain & Ireland from 2010 2020. The include TripAdvisor all restaurants that were awarded stars during these 11 years, large pool...
Download This Paper Open PDF in Browser Add to My Library Share: Permalink Using these links will ensure access this page indefinitely Copy URL DOI
Download This Paper Open PDF in Browser Add to My Library Share: Permalink Using these links will ensure access this page indefinitely Copy URL DOI
Recent investigations described the use of urinary matrix remodeling-associated protein 5 (MXRA5) as a novel biomarker for kidney impairment in setting chronic disease. In this study, we aimed to evaluate possible clinical application MXRA5 useful noninvasive marker urine from affected renal pelvis and bladder children with ureteropelvic junction obstruction (UPJO). This prospective cohort study enrolled patients aged <12 months prenatally diagnosed unilateral UPJO who received dismembered...
Television (TV), the predominant advertising medium, is being transformed by micro-targeting capabilities of set-top boxes (STBs). Accordingly, this paper uses a proprietary, household-level, single source data set to develop second-by-second show and advertisement viewing model, using approach forecast consumers’ exposure downstream consequences for sales.
In many digital contexts such as online news and e-tailing with new users items, recommendation systems face several challenges: i) how to make initial recommendations little or no response history (i.e., cold-start problem), ii) learn user preferences on items (test learn), iii) scale across myriad demographics attributes. While accommodate aspects of these challenges, few if any address all. This paper introduces a Collaborative Filtering (CF) Multi-armed Bandit (B) Attributes (A) system...
Previous research shows that the impact of regulations on innovation is mixed, prompting us to examine this issue within realm environmental regulations. After Chinese government selected 117 key cities for pollution control in 2012, it introduced stringent emissions standards nationwide promote green 2016. By considering introduction as an event and those treatment group (with all other group), we develop a difference-in-differences approach with series robustness checks effects new set...
The fibroblast (FB), cardiomyocyte (CM), and vascular endothelial cell (Vas_EC) are the three major types in heart, yet their relationships during development largely unexplored. To address this gap, we employed RNA staining of FB marker gene
Download This Paper Open PDF in Browser Add to My Library Share: Permalink Using these links will ensure access this page indefinitely Copy URL DOI
Download This Paper Open PDF in Browser Add to My Library Share: Permalink Using these links will ensure access this page indefinitely Copy URL DOI
By analyzing a large-scale transactions data set from grocery chain, we observe that customers arrive randomly at non-stationary rate, they usually purchase multiple products together and in continuous quantity, are often out of stock. However, extant demand estimation methods do not consider these features simultaneously, which can result biased lead to supply mismatch. In this study, develop unified parsimonious model for structural fill gap. First, incorporate stockout into the framework...