Abstract
Motivated by the recently launched mobile data trading markets (e.g., China Mobile Hong Kong's 2nd exChange Market), in this paper we study the mobile data trading problem under the future data demand uncertainty. We introduce a brokerage-based market, where sellers and buyers propose their selling and buying quantities, respectively, to the trading platform that matches the market supply and demand. To understand the users' realistic trading behaviors, a prospect theory (PT) model from behavioral economics is proposed, which includes the widely adopted expected utility theory (EUT) as a special case. Although the PT modeling leads to a challenging non-convex optimization problem, the optimal solution can be characterized by exploiting the unimodal structure of the objective function. Building upon our analysis, we design an algorithm to help estimate the user's risk preference and provide trading recommendations dynamically, considering the latest market and usage information. It is shown in our simulation that the risk preferences have a significant impact on the user's decision and outcome: a risk-averse dominant user can guarantee a higher minimum profit in the trading, while a risk-seeking dominant user can achieve a higher maximum profit. By comparing with the EUT benchmark, it is shown that a PT user with a low reference point is more willing to buy mobile data. Moreover, when the probability of high future data demand is low, a PT user is more willing to buy mobile data due to the probability distortion comparing with an EUT user.
Introduction: The Perishable Commodity Problem
Most mobile data plans follow a "use it or lose it" model: users pay for a monthly quota, and any unused data vanishes at the end of the billing cycle. Conversely, exceeding the limit results in steep overage fees. To address this inefficiency, secondary markets like China Mobile Hong Kong’s "2nd exChange Market" (2CM) have emerged, allowing users to trade their data quotas.
However, trading mobile data is complicated by demand uncertainty. When a user decides to sell data mid-month, they do not yet know their exact requirements for the remaining period. Traditional economic models assume users are "Expected Utility Theory" (EUT) maximizers—rational agents who weigh outcomes purely by their mathematical probability. In reality, human psychology is far more complex; we fear losses more than we value equivalent gains and often misjudge probabilities. This paper bridges the gap between theoretical rationality and actual human behavior by applying Prospect Theory (PT) to mobile data trading.
The Trading Framework
The authors formulate a two-stage optimization problem for a user interacting with a brokerage-based market. In Stage I, the user decides whether to act as a buyer or a seller based on market prices and their own usage profile. In Stage II, they determine the specific quantity of gigabytes (GB) to trade.
Modeling Human Behavior with Prospect Theory
To capture realistic trading behaviors, the study replaces the standard EUT model with Prospect Theory (PT), which incorporates three psychological pillars:
- Reference Points: Users evaluate outcomes as "gains" or "losses" relative to a mental baseline (e.g., expecting the best-case vs. worst-case usage scenario).
- S-Shaped Value Function: Humans are generally risk-averse regarding gains (the curve flattens as gains increase) but risk-seeking when it comes to avoiding losses (the curve is steeper for losses).
- Probability Distortion: We tend to overweight low-probability events (like a rare data spike) and underweight high-probability ones.
While PT makes the mathematics more difficult—resulting in a non-convex optimization problem—the authors demonstrate that the objective function possesses a unimodal structure. This allows for the identification of globally optimal trading quantities by analyzing specific sub-intervals of the feasible trading range.
Key Insights and Results
The study utilized market data from the CMHK 2CM platform and a survey of over 50 users to ground its simulations. Key findings include:
- The "Loss Aversion" Flip: A user’s decision to buy or sell changes drastically based on their reference point. If a user focuses on a "worst-case" high-demand scenario, they become much more willing to buy data to avoid the "pain" of overage fees, even if the expected value suggests otherwise.
- Distorted Reality: Because humans overweight low probabilities, a PT user is more likely to buy "insurance" data when the chance of high demand is low, compared to a "rational" EUT user.
- Risk Profiles: Risk-averse users prioritize guaranteeing a higher minimum profit (avoiding the worst outcomes), while risk-seeking users achieve higher maximum profits by gambling on their future usage.
- Algorithm Performance: The proposed algorithm successfully estimates a user's risk parameters by observing their "indifference prices." In simulations across 1,000 billing cycles, the PT-based algorithm reduced net payments by approximately 50% compared to a "no trading" strategy.
Conclusion
While a risk-neutral EUT model might achieve the highest average profit in a purely mathematical sense, the PT model better captures the behavior of real people who prioritize avoiding "disaster" (overages) or "waste" (unused data). Future work may expand this from a "myopic" single-period model to a dynamic multi-period approach, and explore how market equilibrium shifts when all participants utilize behavioral algorithms.
Summary updated