Liangfei Qiu

Effects of Social Networks on Prediction Markets: Examination in a Controlled Experiment

This paper examines the effect of a social network on prediction markets using a controlled laboratory experiment that allows us to identify causal relationships between a social network and the performance of an individual participant, as well as the performance of the prediction market as a whole. Through a randomized experiment, we first confirm the theoretical predictions that participants with more social connections are less likely to invest in information acquisition from outside information sources but perform significantly better than other participants in prediction markets. We further show that when the cost of information acquisition is low, a social-network-embedded prediction market outperforms a non-networked prediction market. We also find strong support for peer effects in prediction accuracy among participants. These results have direct managerial implications for the business practice of prediction markets and are critical to understanding how to use social networks to improve the performance of prediction markets.

The Impact of Social Network Structures on Prediction Market Accuracy in the Presence of Insider Information

This paper examines the effects of social network structures on prediction market accuracy in the presence of insider information through a randomized laboratory experiment. In the experiment, insider information is operationalized as signals on the state of nature with high precision. Motivated by the literature on insider information in the context of financial markets, we test and confirm two characterizations of insider information in the context of prediction markets: abnormal performance and less diffusion. Experimental results suggest that a more balanced social network structure is crucial to the success of prediction markets, whereas network structures akin to star networks are ill suited to prediction markets. As compared with other network structures, insider information has less positive effects on prediction market accuracy in star networks. We also find that the bias of the public information has a larger negative effect on prediction market accuracy in star networks.

A Friend Like Me: Modeling Network Formation in a Location-Based Social Network

Abstract This article studies the strategic network formation in a location-based social network. We build an empirical model of social link creation that incorporates individual characteristics and pairwise user similarities. Specifically, we define four user proximity measures from biography, geography, mobility, and short messages. To construct proximity from unstructured text information, we build topic models using Latent Dirichlet Allocation. Using Gowalla data with 385,306 users, 3 million locations, and 35 million check-in records, we empirically estimate the model to find evidence on the homophily effect on network formation. To cope with possible endogeneity issues, we use exogenous weather shocks as our instrumental variables and find the empirical results are robust: network formation decisions are significantly affected by our proximity measures.

When Cellular Capacity Meets WiFi Hotspots: A Smart Auction System for Mobile Data Offloading

The surge of social networking and video streaming on the go has led to the explosion of mobile data traffic. To minimize congestion costs for under-served demand (e.g., Dissatisfied customers, or churn), the cellular service provider is willing to pay WiFi hotspots to serve the demand that exceeds capacity. In the present study, we propose an optimal procurement mechanism with contingent contracts for cellular service providers to leverage the advantages of both cellular and WiFi resources. We show the procedure of computing the optimal procurement mechanism with a tight integration of economics and computational technology. Simulation results show that the proposed procurement mechanism significantly outperforms the standard Vickrey-Clarke-Groves (VCG) auction in terms of the cellular service providers expected payoff.

Understanding Voluntary Knowledge Provision and Content Contribution through a Social Media-Based Prediction Market: A Field Experiment

The performance of prediction markets depends crucially on the quality of user contribution. A social media-based prediction market can utilize aspects of social effects to improve users’ contribution quality. Drawing upon literature from diverse areas such as prediction markets, knowledge contribution, public goods provision, and user generated content, we examine the causal effect of social audience size and online endorsement on prediction market participants’ prediction accuracy through a randomized field experiment. By conducting a comprehensive treatment effect analysis, we estimate both the average treatment effect (ATE) and quantile treatment effect (QTE) using the difference-in-differences method. Our empirical results on ATE show that an increase in audience size leads to an increase in prediction accuracy, and that an increase in online endorsement also leads to prediction improvements. Interestingly, we find that quantile treatment effects are heterogeneous: users of intermediate prediction ability respond most positively to an increase in social audience size and online endorsement. These findings suggest that corporate prediction markets can target people of intermediate abilities to obtain the most significant prediction improvement.

Strategic Network Formation in a Location-Based Social Network: A Topic Modeling Approach

This paper studies strategic network formation in a location-based social network. We build a structural model of social link creation that incorporates individual characteristics and pairwise user similarities. Specifically, we define four user proximity measures from biography, geography, mobility, and short messages. To construct proximity from unstructured text information, we build topic models using latent Dirichlet allocation. Using Gowalla data with 385,306 users, three million locations, and 35 million check-in records, we empirically estimate the structural model to find evidence on the homophily effect in network formation.

Detecting Review Manipulation on Online Platforms with Hierarchical Supervised Learning

Abstract Opinion spammers exploit consumer trust by posting false or deceptive reviews that may have a negative impact on both consumers and businesses. These dishonest posts are difficult to detect because of complex interactions between several user characteristics, such as review velocity, volume, and variety. We propose a novel hierarchical supervised-learning approach to increase the likelihood of detecting anomalies by analyzing several user features and then characterizing their collective behavior in a unified manner. Specifically, we model user characteristics and interactions among them as univariate and multivariate distributions. We then stack these distributions using several supervised-learning techniques, such as logistic regression, support vector machine, and k-nearest neighbors yielding robust meta-classifiers. We perform a detailed evaluation of methods and then develop empirical insights. This approach is of interest to online business platforms because it can help reduce false reviews and increase consumer confidence in the credibility of their online information. Our study contributes to the literature by incorporating distributional aspects of features in machine-learning techniques, which can improve the performance of fake reviewer detection on digital platforms.

Exit, Voice, and Response on Digital Platforms: An Empirical Investigation of Online Management Response Strategies

In the past decade, we have witnessed the growing importance of management responses to online reviews on digital platforms. In this study, we examine the impact of online management responses on business performance and their spillover effect on nearby businesses. By adopting multiple causal identification strategies to address the issue of self-selected responses, we find that the responses by a business owner play a significant role in the performance of the focal business as well as in the performance of the nearby businesses. We observe that, in general, the launch of the new management response feature benefits businesses. What is more interesting is that the benefit is not observed in a consistent manner across all businesses. Only the businesses that choose to use the management response feature observe increases in check-ins. On the other hand, the businesses that are unaware of the management response feature launch on digital platforms, or are aware but choose not to use the management response feature, tend to remain at a disadvantage. Interestingly, we also uncover that the spillover effect externality of online management responses on the nearby businesses crucially depends on whether the focal business and the nearby businesses are in direct competition. Furthermore, we identify conditions under which business owners are more likely to respond to consumer comments. Our findings have direct implications for both business owners and digital platforms: Digital platforms can help businesses develop the right engagement strategies by taking the online response strategies of the nearby businesses into account. The online appendix is available at https://doi.org/10.1287/isre.2017.0749 .

Repeated interactions versus social ties: quantifying the economic value of trust, forgiveness, and reputation using a field experiment

The growing importance of online social networks provides fertile ground for researchers seeking to gain a deeper understanding of fundamental constructs of human behavior, such as trust and forgiveness, and their linkage to social ties. Through a field experiment that uses data from the Facebook API to measure social ties that connect our subjects, we separate forward-looking instrumental trust from static intrinsic trust and show that the level of instrumental trust and forgiveness, and the effect of forgiveness on deterring future defections, crucially depend on the strength of social ties. We find that the level of trust under social repeated play is greater than the level of trust under anonymous repeated play, which in turn is greater than the level of trust under anonymous one shot games. We also uncover forgiveness as a key mechanism that facilitates the cooperative equilibrium being more stable in the presence of social ties: If the trading partners are socially connected, the equilibrium is more likely to return to the original cooperative one after small disturbances.

'Hidden Profiles' in Corporate Prediction Markets: The Impact of Public Information Precision and Social Interactions

Recently, large companies are experimenting with corporate prediction markets run among their employees. In the present study, we develop an analytical model to analyze the effects of information precision and social interactions on prediction market performance. We find that increased precision of public information is not always beneficial to the prediction market accuracy because of the “hidden profiles” effect: the information-aggregation mechanism places a larger-than-efficient weight on existing public information. We show that a socially embedded prediction market with information sharing among participants may help correct such inefficiency and improve the prediction market performance. We also identify conditions under which increased precision of public information is detrimental in a non-networked prediction market and in a socially embedded prediction market. These results should be of interest to practitioners as the managerial implications highlight the detrimental effect of public information and the role of social networking among employees in a corporate prediction market.