arXiv subjects

Gerardo Berbeglia

Publications and source records attributed to Gerardo Berbeglia.

At least 19 recordsLinked to original sources

arXiv2026

Estimation, Prediction, and Assortment Optimization for Markov Chain Choice Models with Panel Data

We propose a framework for the Markov chain (MC) choice model with panel data, including parameter estimation, personalized choice prediction, and personalized assortment optimization. In contrast to the traditional setting, which assumes that each transaction is independently drawn from a random utility model, our framework accounts for dependencies among transactions for the same customer in historical data, captured by partial-ordering preference information. To the best of our knowledge, our framework initiates the study of choice modeling with panel data under MC. As our primary result, we propose novel expectation-maximization (EM) algorithms for MC parameter estimation by incorporating partial-ordering-based customer preference information. On synthetic datasets and the sushi dataset, our EM algorithms outperform the traditional EM algorithm of Simsek and Topaloglu (Operations Research, 66, 2018) and multinomial-logit-based partial-order benchmarks adapted from Jagabathula and Vulcano (Management Science, 64, 2018). As our secondary contribution, we present hardness and computational results for conditional choice prediction and assortment optimization problems. These results complement our estimation framework and clarify the computational landscape of conditional choice and assortment optimization, which may be of independent interest.

cs.LG↗

arXiv2026

Modern column generation for estimating single- and multi-purchase ranked list choice models

This paper studies the estimation of ranked-list discrete choice models with single and multiple purchases. In this setting, each consumer type is characterized by a ranking over a subset of products and a desired number of purchases, and the estimation task is to identify the set of consumer types and their probabilities that best explain the observed transactional data. This problem is computationally challenging due to the exponential number of possible consumer types and becomes more difficult when multiple purchases are allowed. We propose a column generation framework for this problem. Our main contribution is a dynamic programming algorithm for the column generation subproblem. This subproblem generalizes the linear ordering problem and incorporates acceleration techniques to improve computational efficiency. To the best of our knowledge, this is the first dynamic programming-based approach for generating consumer types in non-parametric models. The proposed framework supports multiple model variants with minor modifications. Computational experiments on synthetic and real data show substantial speedups over existing methods while maintaining high solution quality, and demonstrate effectiveness in both estimation and assortment optimization.

cs.DS↗

arXiv2025

The generalized stochastic preference choice model

We propose a new discrete choice model, called the generalized stochastic preference (GSP) model, that incorporates non-rationality into the stochastic preference (SP) choice model, also known as the rank-based model. Our model can capture several context-dependent choice behaviors that cannot be represented by any SP model, such as the well-documented compromise and attraction effects, while still including the SP model as a special case. The GSP model is defined as a distribution over consumer types, where each type extends the choice behavior of rational types in the SP model. We build on existing methods for estimating the SP model and propose an iterative estimation algorithm for the GSP model that finds new types by solving an integer linear program in each iteration. We further show that our proposed notion of non-rationality can be incorporated into other choice models, like the random utility maximization (RUM) model class as well as any of its subclasses. As a concrete example, we introduce the non-rational extension of the classical MNL model, which we term the generalized MNL (GMNL) model and present an efficient expectation-maximization (EM) algorithm for estimating it. \av{For the GSP model, we demonstrate that the worst-case performance guarantee of revenue-ordered assortments is significantly worse than for the SP model. For the GMNL model, we establish that assortment optimization with totally unimodular constraints is NP-hard to approximate to within a factor of $O(n^{1-ε})$ for any $ε> 0$, where $n$ is the number of products.} Finally, numerical evaluation on synthetic and real choice data shows that the GMNL and GSP models can outperform their rational counterparts in out-of-sample prediction accuracy.

cs.GT↗

arXiv2024

Approximate Resolution of Stochastic Choice-based Discrete Planning

Stochastic choice-based discrete planning is a broad class of decision-making problems characterized by a sequential decision-making process involving a planner and a group of customers. The firm or planner first decides a subset of options to offer to the customers, who, in turn, make selections based on their utilities of those options. This problem has extensive applications in many areas, including assortment planning, product line design, and facility location. A key feature of these problems is that the firm cannot fully observe the customers' utilities or preferences, which results in intrinsic and idiosyncratic uncertainties. Most works in the literature have studied a specific type of uncertainty, resulting in customized decision models that are subsequently tackled using ad-hoc algorithms designed to exploit the specific model structure. In this paper we propose a modeling framework capable of solving this family of sequential problems that works for a large variety of uncertainties. We then leverage an approximation scheme and develop an adaptable mixed-integer linear programming method. To speed up the solution process, we further develop an efficient decomposition approach. We show that our solution framework can yield solutions proven to be (near-)optimal for a broad class of problems. We illustrate this by applying our approach to three classical application problems: constrained assortment optimization and two facility location problems. Through extensive computational experiments, we demonstrate the performance of our approach in terms of both solution quality and computational speed, and provide computational insights. In particular, when we use our method to solve the constrained assortment optimization problem under the Exponomial choice model, it improves the state-of-the-art.

math.OC↗

arXiv2024

The Limits of Personalization in Assortment Optimization

To study the limits of personalization, we introduce the notion of a clairvoyant firm that can read the mind of consumers and sell them the highest revenue product that they are willing to buy. We show how to compute the expected revenue of the clairvoyant firm for a class of rational discrete choice models, and develop prophet-type inequalities that provide performance guarantees for the expected revenue of the traditional assortment optimization firm (a TAOP firm) relative to the clairvoyant firm, and therefore to any effort to personalize assortments. In particular, we show that the expected revenue of the clairvoyant firm cannot exceed twice the expected revenue of the TAOP for the RCS model, the MNL, the GAM and the Nested Logit Model. On the other hand, there are random utility models for which personalized assortments can earn up to $n$ times more than a TAOP firm, where $n$ is the number of products. Our numerical studies indicate that when the mean utilities of the products are heterogeneous among consumer types, and the variance of the utilities is small, firms can gain substantial benefits from personalized assortments. We support these observations, and others, with theoretical findings. While the consumers surplus can potentially be larger under personalized assortments, clairvoyant firms with pricing power can extract all surplus, and earn arbitrarily more than traditional firms that optimize over prices but do not personalize them. For the price-aware MNL, however, a clairvoyant firm can earn at most $e$ times more than a traditional firm.

econ.TH↗

arXiv2021

Bounds and Heuristics for Multi-Product Personalized Pricing

We present tight bounds and heuristics for personalized, multi-product pricing problems. Under mild conditions we show that the best price in the direction of a positive vector results in profits that are guaranteed to be at least as large as a fraction of the profits from optimal personalized pricing. For unconstrained problems, the fraction depends on the factor and on optimal price vectors for the different customer types. For constrained problems the factor depends on the factor and a ratio of the constraints. Using a factor vector with equal components results in uniform pricing and has exceedingly mild sufficient conditions for the bound to hold. A robust factor is presented that achieves the best possible performance guarantee. As an application, our model yields a tight lower-bound on the performance of linear pricing relative to optimal personalized non-linear pricing, and suggests effective non-linear price heuristics relative to personalized solutions. Additionally, our model provides guarantees for simple strategies such as bundle-size pricing and component-pricing with respect to optimal personalized mixed bundle pricing. Heuristics to cluster customer types are also developed with the goal of improving performance by allowing each cluster to price along its own factor. Numerical results are presented for a variety of demand models that illustrate the tradeoffs between using the economic factor and the robust factor for each cluster, as well as the tradeoffs between using a clustering heuristic with a worst case performance of two and a machine learning clustering algorithm. In our experiments economically motivated factors coupled with machine learning clustering heuristics performed best.

econ.TH↗

arXiv2021

Market Segmentation in Online Platforms

This paper studies ranking policies in a stylized trial-offer marketplace model, in which a single firm offers products and has consumers with heterogeneous preferences. Consumer trials are influenced by past purchases and the ranking of each product. The platform owner needs to devise a ranking policy to display the products to maximize the number of purchases in the long run. The model proposed attempts to understand the impact of market segmentation in a trial-offer market with social influence. In our model, consumer choices are based on a very general choice model known as the mixed MNL. We analyze the long-term dynamics of this highly complex stochastic model and we quantify the expected benefits of market segmentation. When past purchases are displayed, consumer heterogeneity makes buyers try the sub-optimal products, reducing the overall sales rate. We show that consumer heterogeneity makes the ranking problem NP-hard. We then analyze the benefits of market segmentation. We find tight bounds to the expected benefits of offering a distinct ranking to each consumer segment. Finally, we show that the market segmentation strategy always benefits from social influence when the average quality ranking is used. One of the managerial implications is that the firm is better off using an aggregate ranking policy when the variety of consumer preference is limited, but it should perform a market segmentation policy when consumers are highly heterogeneous. We also show that this result is robust to relatively small consumer classification mistakes; when these are large, an aggregate ranking is preferred.

cs.SI↗

arXiv2021

The Refined Assortment Optimization Problem

We introduce the refined assortment optimization problem where a firm may decide to make some of its products harder to get instead of making them unavailable as in the traditional assortment optimization problem. Airlines, for example, offer fares with severe restrictions rather than making them unavailable. This is a more subtle way of handling the trade-off between demand induction and demand cannibalization. For the latent class MNL model, a firm that engages in refined assortment optimization can make up to $\min(n,m)$ times more than one that insists on traditional assortment optimization, where $n$ is the number of products and $m$ the number of customer types. Surprisingly, the revenue-ordered assortment heuristic has the same performance guarantees relative to {\em personalized} refined assortment optimization as it does to traditional assortment optimization. Based on this finding, we construct refinements of the revenue-order heuristic and measure their improved performance relative to the revenue-ordered assortment and the optimal traditional assortment optimization problem. We also provide tight bounds on the ratio of the expected revenues for the refined versus the traditional assortment optimization for some well known discrete choice models.

econ.TH↗

arXiv2020

Pricing under a multinomial logit model with non linear network effects

We study the problem of pricing under a Multinomial Logit model where we incorporate network effects over the consumer's decisions. We analyse both cases, when sellers compete or collaborate. In particular, we pay special attention to the overall expected revenue and how the behaviour of the no purchase option is affected under variations of a network effect parameter. Where for example we prove that the market share for the no purchase option, is decreasing in terms of the value of the network effect, meaning that stronger communication among costumers increases the expected amount of sales. We also analyse how the customer's utility is altered when network effects are incorporated into the market, comparing the cases where both competitive and monopolistic prices are displayed. We use tools from stochastic approximation algorithms to prove that the probability of purchasing the available products converges to a unique stationary distribution. We model that the sellers can use this stationary distribution to establish their strategies. Finding that under those settings, a pure Nash Equilibrium represents the pricing strategies in the case of competition, and an optimal (that maximises the total revenue) fixed price characterise the case of collaboration.

cs.GT↗

arXiv2019

Assortment and Price Optimization Under the Two-Stage Luce model

This paper studies assortment and pricing optimization problems under the Two-Stage Luce model (2SLM), a discrete choice model introduced by Echenique and Saito (2018) that generalizes the multinomial logit model (MNL). The model employs an utility function as in the the MNL, and a dominance relation between products. When consumers are offered an assortmentS , they first discard all dominated products in S and then select one of the remaining products using the standard MNL. This model may violate the regularity condition, which states that the probability of choosing a product cannot increase if the offer set is enlarged. Therefore, the 2SLM falls outside the large family of discrete choice models based on random utility which contains almost all choice models studied in revenue management. We prove that the assortment problem under the 2SLM is polynomial-time solvable. Moreover, we show that the capacitated assortment optimization problem is NP-hard and but it admits polynomial-time algorithms for the relevant special cases cases where (1) the dominance relation is attractiveness-correlated and (2) its transitive reduction is a forest. The proofs exploit a strong connection between assortments under the 2SLM and independent sets in comparability graphs. Finally, we study the associated joint pricing and assortment problem under this model. First, we show that well known optimal pricing policy for the MNL can be arbitrarily bad. Our main result in this section is the development of an efficient algorithm for this pricing problem. The resulting optimal pricing strategy is simple to describe: it assigns the same price for all products, except for the one with the highest attractiveness and as well as for the one with the lowest attractiveness.

cs.DM↗

arXiv2019

Assortment optimisation under a general discrete choice model: A tight analysis of revenue-ordered assortments

The assortment problem in revenue management is the problem of deciding which subset of products to offer to consumers in order to maximise revenue. A simple and natural strategy is to select the best assortment out of all those that are constructed by fixing a threshold revenue $π$ and then choosing all products with revenue at least $π$. This is known as the revenue-ordered assortments strategy. In this paper we study the approximation guarantees provided by revenue-ordered assortments when customers are rational in the following sense: the probability of selecting a specific product from the set being offered cannot increase if the set is enlarged. This rationality assumption, known as regularity, is satisfied by almost all discrete choice models considered in the revenue management and choice theory literature, and in particular by random utility models. The bounds we obtain are tight and improve on recent results in that direction, such as for the Mixed Multinomial Logit model by Rusmevichientong et al. (2014). An appealing feature of our analysis is its simplicity, as it relies only on the regularity condition. We also draw a connection between assortment optimisation and two pricing problems called unit demand envy-free pricing and Stackelberg minimum spanning tree: These problems can be restated as assortment problems under discrete choice models satisfying the regularity condition, and moreover revenue-ordered assortments correspond then to the well-studied uniform pricing heuristic. When specialised to that setting, the general bounds we establish for revenue-ordered assortments match and unify the best known results on uniform pricing.

cs.DS↗

arXiv2018

Assortment Optimization under the Sequential Multinomial Logit Model

We study the assortment optimization problem under the Sequential Multinomial Logit (SML), a discrete choice model that generalizes the multinomial logit (MNL). Under the SML model, products are partitioned into two levels, to capture differences in attractiveness, brand awareness and, or visibility of the products in the market. When a consumer is presented with an assortment of products, she first considers products in the first level and, if none of them is purchased, products in the second level are considered. This model is a special case of the Perception-Adjusted Luce Model (PALM) recently proposed by Echenique et al (2018). It can explain many behavioural phenomena such as the attraction, compromise, similarity effects and choice overload which cannot be explained by the MNL model or any discrete choice model based on random utility. In particular, the SML model allows violations to regularity which states that the probability of choosing a product cannot increase if the offer set is enlarged. This paper shows that the seminal concept of revenue-ordered assortment sets, which contain an optimal assortment under the MNL model, can be generalized to the SML model. More precisely, the paper proves that all optimal assortments under the SML are revenue-ordered by level, a natural generalization of revenue-ordered assortments that contains, at most, a quadratic number of assortments. As a corollary, assortment optimization under the SML is polynomial-time solvable. This result is particularly interesting given that the SML model does not satisfy the regularity condition and, therefore, it can explain choice behaviours that cannot be explained by any choice model based on random utility.

cs.DM↗

arXiv2018

Pricing Policies for Selling Indivisible Storable Goods to Strategic Consumers

We study the dynamic pricing problem faced by a monopolistic retailer who sells a storable product to forward-looking consumers. In this framework, the two major pricing policies (or mechanisms) studied in the literature are the preannounced (commitment) pricing policy and the contingent (threat or history dependent) pricing policy. We analyse and compare these pricing policies in the setting where the good can be purchased along a finite time horizon in indivisible atomic quantities. First, we show that, given linear storage costs, the retailer can compute an optimal preannounced pricing policy in polynomial time by solving a dynamic program. Moreover, under such a policy, we show that consumers do not need to store units in order to anticipate price rises. Second, under the contingent pricing policy rather than the preannounced pricing mechanism, (i) prices could be lower, (ii) retailer revenues could be higher, and (iii) consumer surplus could be higher. This result is surprising, in that these three facts are in complete contrast to the case of a retailer selling divisible storable goods Dudine et al. (2006). Third, we quantify exactly how much more profitable a contingent policy could be with respect to a preannounced policy. Specifically, for a market with $N$ consumers, a contingent policy can produce a multiplicative factor of $Ω(\log N)$ more revenues than a preannounced policy, and this bound is tight.

cs.GT↗

arXiv2017

Popularity Signals in Trial-Offer Markets with Social Influence and Position Bias

This paper considers trial-offer markets where consumer preferences are modeled by a multinomial logit with social influence and position bias. The social signal for a product is given by its current market share raised to power r (or equivalently the number of purchases raised to the power of r). The paper shows that, when r is strictly between 0 and 1, and a static position assignment (e.g., a quality ranking) is used, the market converges to a unique equilibrium where the market shares depend only on product quality, not their initial appeals or the early dynamics. When r is greater than 1, the market becomes unpredictable. In many cases, the market goes to a monopoly for some product: Which product becomes a monopoly depends on the initial conditions of the market. These theoretical results are complemented by an agent-based simulation which indicates that convergence is fast when r is between 0 and 1, and that the quality ranking dominates the well-known popularity ranking in terms of market efficiency. These results shed a new light on the role of social influence which is often blamed for unpredictability, inequalities, and inefficiencies in markets. In contrast, this paper shows that, with a proper social signal and position assignment for the products, the market becomes predictable, and inequalities and inefficiencies can be controlled appropriately.

cs.SI↗

arXiv2016

The effect of a finite time horizon in the durable good monopoly problem with atomic consumers

A durable good is a long-lasting good that can be consumed repeatedly over time, and a duropolist is a monopolist in the market of a durable good. In 1972, Ronald Coase conjectured that a duropolist who lacks commitment power cannot sell the good above the competitive price if the time between periods approaches zero. Coase's counterintuitive conjecture was later proven by Gul et al. (1986) under an infinite time horizon model with non-atomic consumers. Remarkably, the situation changes dramatically for atomic consumers and an infinite time horizon. Bagnoli et al. (1989) showed the existence of a subgame-perfect Nash equilibrium where the duropolist extracts all the consumer surplus. Observe that, in these cases, duropoly profits are either arbitrarily smaller or arbitrarily larger than the corresponding static monopoly profits -- the profit a monopolist for an equivalent consumable good could generate. In this paper we show that the result of Bagnoli et al. (1989) is in fact driven by the infinite time horizon. Indeed, we prove that for finite time horizons and atomic agents, in any equilibrium satisfying the standard skimming property, duropoly profits are at most an additive factor more than static monopoly profits. In particular, duropoly profits are always at least static monopoly profits but never exceed twice the static monopoly profits. Finally we show that, for atomic consumers, equilibria may exist that do not satisfy the skimming property. For two time periods, we prove that amongst all equilibria that maximize duropoly profits, at least one of them satisfies the skimming property. We conjecture that this is true for any number of time period.

cs.GT↗

arXiv2016

Trial-Offer Markets with Continuation

Trial-offer markets, where customers can sample a product before deciding whether to buy it, are ubiquitous in the online experience. Their static and dynamic properties are often studied by assuming that consumers follow a multinomial logit model and try exactly one product. In this paper, we study how to generalize existing results to a more realistic setting where consumers can try multiple products. We show that a multinomial logit model with continuation can be reduced to a standard multinomial logit model with different appeal and product qualities. We examine the consequences of this reduction on the performance and predictability of the market, the role of social influence, and the ranking policies.

cs.GT↗

arXiv2016

Interdependent Scheduling Games

We propose a model of interdependent scheduling games in which each player controls a set of services that they schedule independently. A player is free to schedule his own services at any time; however, each of these services only begins to accrue reward for the player when all predecessor services, which may or may not be controlled by the same player, have been activated. This model, where players have interdependent services, is motivated by the problems faced in planning and coordinating large-scale infrastructures, e.g., restoring electricity and gas to residents after a natural disaster or providing medical care in a crisis when different agencies are responsible for the delivery of staff, equipment, and medicine. We undertake a game-theoretic analysis of this setting and in particular consider the issues of welfare maximization, computing best responses, Nash dynamics, and existence and computation of Nash equilibria.

cs.GT↗

arXiv2015

Bargaining Mechanisms for One-Way Games

We introduce one-way games, a framework motivated by applications in large-scale power restoration, humanitarian logistics, and integrated supply-chains. The distinguishable feature of the games is that the payoff of some player is determined only by her own strategy and does not depend on actions taken by other players. We show that the equilibrium outcome in one-way games without payments and the social cost of any ex-post efficient mechanism, can be far from the optimum. We also show that it is impossible to design a Bayes-Nash incentive-compatible mechanism for one-way games that is budget-balanced, individually rational, and efficient. To address this negative result, we propose a privacy-preserving mechanism that is incentive-compatible and budget-balanced, satisfies ex-post individual rationality conditions, and produces an outcome which is more efficient than the equilibrium without payments. The mechanism is based on a single-offer bargaining and we show that a randomized multi-offer extension brings no additional benefit.

cs.GT↗