Discrete Choice Model Of Demand Research Articles

Cheap Tuesdays and the Demand for Cinema

Many movie markets are characterised by extensive uniform pricing practices, hampering the ability to estimate price elasticities of demand. Australia presents a rare exception, with most cinemas offering cheap Tuesday ticket prices. We exploit this feature to estimate a random coefficients discrete choice model of demand for the Sydney region in 2007. We harness an extensive set of film, cinema, and time-dependent characteristics to build a rich demand system. Our results are consistent with a market expansion effect from the practice of discounted Tuesday tickets, and suggest that cinemas could profit from price dispersion by discounts based on observable characteristics.

Merger effects with product complementarity: Evidence from Colombia’s telecommunications

Mergers of firms producing complementary products entail two opposing effects: Lower prices, because they may internalize the impact of the complementarity, and higher prices, because the firms gain the ability to price discriminate. I use Colombian data to analyze mergers between firms providing complementary telecommunication services. I estimate a discrete choice model of demand for bundled fixed-line and mobile internet services, in which the degree of either substitutability or complementarity among products is a parameter of interest. Counterfactual experiments using the estimated model indicate pro-competitive effects of mergers with complements: despite a small increase in the price of standalone goods, consumer surplus increases.

Estimating dynamic discrete choice models with aggregate data: Properties of the inclusive value approximation

We investigate the use of the inclusive value based approach for estimating dynamic discrete choice models of demand with aggregate data. The inclusive value sufficiency (IVS) approach approximates a multi-dimensional state space with a single “sufficient statistic” in order to mitigate the curse of dimensionality and tractability estimate model primitives. Although in widespread use, the conditions under which IVS is appropriate have not been examined. Theoretically, we show that the estimator is biased and inconsistent. We then use Monte Carlo simulations (of a simple model of dynamic durable goods adoption) to demonstrate the degree of bias associated with the inclusive value approximation estimator under an array of parameterizations and data generating processes. In our examination, we show that the estimator performs better when the discount factor is smaller and/or when the price sensitivity of the consumer is larger. Examining how the bias impacts economic quantities of interest, we find that the IVS method under estimates the true long-run own-price elasticities and over estimates the change in profits as prices change. Theses findings highlight the importance of correctly specifying how consumers form expectations. As a result, researchers should consider how to empirically support their assumption for the underlying consumer belief structure.

A Practical Price Optimization Approach for Omnichannel Retailing

Consumers are increasingly navigating across sales channels to maximize the value of their purchase. The existing retail practices of pricing channels either independently or matching competitor prices are unable to achieve the desired profitable coordination between channels. We engaged with three major retailers over two years and developed omnichannel pricing (OCP) solutions in partnership with IBM Commerce to overcome these challenges. We implement an integrated data processing and machine learning framework that enables estimation of location-specific, cross-channel price elasticities and competitive effects. We develop an integrated OCP optimization formulation to profitably coordinate prices for nonperishable products offered across channels and store locations while satisfying practical constraints on volume and price. The resultant optimization formulations for discrete choice demand models are nonconvex and NP-hard, and we prescribe practically efficient mixed-integer programs that can be used to recover (near) optimal solutions. An OCP implementation in two categories for a major retail chain projected a 7% profit lift while preserving the sales volume. This benefit was achieved by lowering online prices and optimally raising and lowering location-specific store prices. This integrated pricing approach allows the retailer to be competitive and preserve market share without aggressively matching the low price of e-tail giants.

Airline schedule padding and consumer choice behavior

Disclosure programs are quality assurance instruments when consumers are imperfectly informed about product quality. This study examines the On-Time Disclosure Rule for airline on-time performance which ranks airlines based on how well their flights arrive/depart on time. The program creates incentives for airlines to pad their flight schedules and artificially improve their reported on-time performance. Using a discrete choice demand model for air travel, we investigate how schedule padding affects consumer preferences. We find that schedule padding is associated with negative consumer welfare effects, reinforcing the notion of an unanticipated gaming behavior by firms that ends up harming consumer welfare.

A strong Lagrangian relaxation for general discrete-choice network revenue management

Discrete-choice network revenue management (DC-NRM) captures both customer behavior and the resource-usage interaction of products, and is appropriate for airline and hotel revenue management, dynamic sales of bundles in advertising, and dynamic assortment optimization in retail. The state-space of the DC-NRM stochastic dynamic program explodes and approximation methods such as the choice deterministic linear program, the affine, and the piecewise-linear approximations have been proposed to approximate it in practice. The affine relaxation (and thereby, its generalization, the piecewise-linear approximation) is intractable even for the simplest choice models such as the multinomial logit (MNL) choice model with a single segment. In this paper we propose a new Lagrangian relaxation method for DC-NRM based on an extended set of multipliers. An attractive feature of our method is that the number of constraints in our formulation scales linearly with the resource capacities. While the number of constraints in our formulation is an order of magnitude smaller that the piecewise-linear approximation (polynomial vs exponential), it obtains a bound that is as tight as the piecewise-linear bound. If we assume that the consideration sets of the different customer segments are small in size—a reasonable modeling tradeoff in many practical applications—our method is an indirect way to obtain the piecewise-linear approximation on large problems effectively. Our results are not specific to a particular functional form (such as MNL), but hold for any discrete-choice model of demand. We show by numerical experiments that our Lagrangian relaxation method can provide substantial improvements over existing benchmark methods, both in terms of tighter upper bounds, as well as revenues from policies based on the relaxation.

Consumer Search in the U.S. Auto Industry: The Value of Dealership Visits

In many markets, consumers visit stores and physically inspect products before making purchase decisions. We view the inspection of a product at a retail location as a search for product fit. We quantify the cost and benefit from searching for product fit using a discrete choice model of demand with optimal sequential search. In these models, the benefit of searching is measured by the standard deviation of the product fit and has, heretofore, been fixed to one in estimation. We show that, with an exogenous search cost shifter, both the cost and benefit of searching can be separately estimated. Our empirical setting is the U.S. automotive market. We assemble a unique data set containing individual-level smartphone geolocation data that inform us about dealership visits. We also obtain information on new vehicle purchases from proprietary DMV registration data. Our exogenous cost shifter is the distance a consumer must travel to visit a dealership. Our results show that the benefit provided by dealerships to consumers is substantial. Within our empirical context, failure to estimate the standard deviation of the product fit leads to biased search cost and consumer surplus estimates and to inaccurate predictions regarding consumers' number of searches and effects of at-home test drive programs.

Transit-oriented autonomous vehicle operation with integrated demand-supply interaction

Autonomous vehicles (AVs) represent potentially disruptive and innovative changes to public transportation (PT) systems. However, the exact interplay between AV and PT is understudied in existing research. This paper proposes a systematic approach to the design, simulation, and evaluation of integrated autonomous vehicle and public transportation (AV + PT) systems. Two features distinguish this research from the state of the art in the literature: the first is the transit-oriented AV operation with the purpose of supporting existing PT modes; the second is the explicit modeling of the interaction between demand and supply.We highlight the transit-orientation by identifying the synergistic opportunities between AV and PT, which makes AVs more acceptable to all the stakeholders and respects the social-purpose considerations such as maintaining service availability and ensuring equity. Specifically, AV is designed to serve first-mile connections to rail stations and provide efficient shared mobility in low-density suburban areas. The interaction between demand and supply is modeled using a set of system dynamics equations and solved as a fixed-point problem through an iterative simulation procedure. We develop an agent-based simulation platform of service and a discrete choice model of demand as two subproblems. Using a feedback loop between supply and demand, we capture the interaction between the decisions of the service operator and those of the travelers and model the choices of both parties. Considering uncertainties in demand prediction and stochasticity in simulation, we also evaluate the robustness of our fixed-point solution and demonstrate the convergence of the proposed method empirically.We test our approach in a major European city, simulating scenarios with various fleet sizes, vehicle capacities, fare schemes, and hailing strategies such as in-advance requests. Scenarios are evaluated from the perspectives of passengers, AV operators, PT operators, and urban mobility system. Results show the trade off between the level of service and the operational cost, providing insight for fleet sizing to reach the optimal balance. Our simulated experiments show that encouraging ride-sharing, allowing in-advance requests, and combining fare with transit help enable service integration and encourage sustainable travel. Both the transit-oriented AV operation and the demand-supply interaction are essential components for defining and assessing the roles of the AV technology in our future transportation systems, especially those with ample and robust transit networks.

How Rewarding is the Reward? Demand Estimation of Crowdfunding Platforms

Reward-based crowdfunding platforms are best known to facilitate the fund-raising process for entrepreneurs and small businesses. Entrepreneurs seek money from the backers to kick off their projects in exchange for rewards. While the competition among the projects to get more supporters grow, many projects fail to reach their fund-raising target. We use a structural demand estimation to understand the role of different aspects of reward scheme design and pricing. More specifically, we apply an aggregate level discrete choice demand model on Kickstarter projects. We characterize the features of the projects by utilizing the document embedding vectors from natural language processing methods. We treat the endogeneity of the price by applying latent instrument variables and finite mixture model for price. The results show that the price coefficient is biased towards zero in the absence of endogeneity treatment. On the reward scheme design, we show low-level rewards cannibalize the more expensive ones. We also find that low-level rewards have the least contribution in customer welfare. Campaign creators and online platforms benefit from the insights of the model for reward scheme design and pricing.

Estimation of random coefficients logit demand models with interactive fixed effects

We extend the Berry, Levinsohn and Pakes (BLP, 1995) random coefficients discrete-choice demand model, which underlies much recent empirical work in IO. We add interactive fixed effects in the form of a factor structure on the unobserved product characteristics. The interactive fixed effects can be arbitrarily correlated with the observed product characteristics (including price), which accommodate endogeneity and, at the same time, capture strong persistence in market shares across products and markets. We propose a two-step least squares-minimum distance (LS-MD) procedure to calculate the estimator. Our estimator is easy to compute, and Monte Carlo simulations show that it performs well. We consider an empirical illustration to US automobile demand.

Quantity-Surcharged Larger Package Sales As Rationally Inattentive Consumers' Choice

Conventional discrete-choice models of demand with fully informed consumers mechanically predict an increase in consumer welfare whenever more alternatives are added to the choice set. When consumers face incomplete information about choice alternatives, the addition of a new alternative could reduce the chance of selecting a superior option, creating ambiguous ex post welfare implications. This paper develops a novel empirical framework of discrete choice and consumer-welfare evaluation based on rational inattention, in which consumers form subjective prior beliefs about products and engage in costly product research to resolve some of the uncertainty prior to purchase. A case study on the introduction of Tide Pods laundry detergent into the market finds the launch of the new product reduced average consumer welfare. Counterfactual experiments suggest Tide would need to lower its Pods price by 6%-7% to restore the average consumer’s pre-launch welfare, and lowering the cost of acquiring information would lead consumers to switch away from choosing the highly priced new Pods products.

A latent-class discrete-choice model of demand for economics electives: a case study on the challenge to increase enrollment

ABSTRACTIncreasing enrollment in undergraduate and graduate economics courses that focus on the natural world (natural resource, environmental, energy, and mineral economics) is important for preparing a work force capable of addressing complex environmental issues. This is especially challenging for smaller universities that want to offer diverse electives in economics. This study seeks to obtain information on the effects of course attributes and student characteristics on preferences for and probabilities to enroll in economics courses offered at Michigan Technological University. A discrete-choice experiment is used to elicit 723 students’ stated preferences for eight different course attributes. A latent-class model is used to probabilistically group students into different preference classes. The cost-effective methods demonstrated in this study can be applied to any courses in any fields to inform course-level marketing efforts aimed at bolstering enrollment while addressing administrative budget constraints.

How Well Do Structural Demand Models Work? Counterfactual Predictions in School Choice

Discrete choice demand models are widely used for counterfactual policy simulations, yet their out-of-sample performance is rarely assessed. This paper uses a large-scale policy change in Boston to investigate the performance of discrete choice models of school demand. In 2013, Boston Public Schools considered several new choice plans that differ in where applicants can apply. At the request of the mayor and district, we forecast the alternatives' effects by estimating discrete choice models. This work led to the adoption of a plan which significantly altered choice sets for thousands of applicants. Pathak and Shi (2014) update forecasts prior to the policy change and describe prediction targets involving access, travel, and unassigned students. Here, we assess how well these ex ante counterfactual predictions compare to actual outcome under the new choice sets. We find that a simple ad hoc model performs as well as the more complicated structural choice models for one of the two grades we examine. However, the structural models' inconsistent performance is largely due to prediction errors in applicant characteristics, which are auxiliary inputs. Once we condition on the actual applicant characteristics, the structural choice models outperform the ad hoc alternative in predicting both choice patterns and policy relevant outcomes. Moreover, refitting the models using the new choice data does not significantly improve their prediction accuracy, suggesting that the choice models are indeed “structural.” Our findings show that structural demand models can effectively predict counterfactual outcomes, as long there are accurate forecasts about auxiliary input variables.

Demand and Consumer Surplus in the On-Demand Economy: The Case of Ride Sharing

Despite the popularity of ridesharing, there is limited empirical evidence on how ridesharing activities differ across regions with different levels of accessibility and the implication for consumers. In this paper, we study the market for rides across New York City neighborhoods. We construct a novel data set that contains massive API queries on route-specific estimates of pricing, wait time, and travel time of Uber, Lyft, and the public transit. After linking this data with actual trip records of taxis, Uber, and Lyft, we document a strong pattern that ridesharing has a larger market share relative to taxis in neighborhoods with lower accessibility, defined either in terms of geographic to Midtown Manhattan or economic distance to job opportunities. Next, we estimate a discrete-choice model of demand for rides and interpret the geography of ridesharing through the lens of the model. We find that consumer surplus from ridesharing varies drastically across geography: passengers that are 5 to 15 miles (resp. more than 15 miles) from Midtown experience a 60% (resp. 19%) larger consumer surplus relative to passengers that are within 5 miles from Midtown. Additionally, over half of these gains comes from reduced wait time. We discuss the implications of the distributional results for policy makers.

Airline on-time performance and its effects on consumer choice behavior

Perhaps, owing to the unavailability of delay data on passengers’ flight routing, a challenge in air travel demand estimation is the examination of consumer response to flight delay within an origin-destination framework. An interesting but unanswered question is: how does air travel delay at various airports along a given itinerary affect consumer choice behavior? Using a methodology that enables us to match airline on-time performance data to passenger itinerary and estimating a discrete choice demand model, this paper finds that increases in flight delay negatively impact the likelihood of choosing a product and that welfare costs of delay to consumers may be substantial. These results are robust across different measures of on-time performance and irrespective of market hauls.

Counterfactual Estimation in Semiparametric Multinomial Choice Models

We show how to construct bounds on counterfactual choice probabilities in semiparametric discrete-choice models. Our procedure is based on cyclic monotonicity, a convex-analytic property of the random utility discrete-choice model. These bounds are useful for typical counterfactual exercises in aggregate discrete-choice demand models. In our semiparametric approach, we do not specify the parametric distribution for the utility shocks, thus accommodating a wide variety of substitution patterns among alternatives. Computation of the counterfactual bounds is a tractable linear programming problem. We illustrate our approach in a series of Monte Carlo simulations and an empirical application using scanner data.

Are Generalized Extreme Value Class Discrete Choice Demand Models Rationalizable as a Representative Consumer's Behavior?

In general, they are not. Most discrete choice demand systems generated from McFadden’s Generalized Extreme Value (GEV) class fail to satisfy the Slutsky symmetry whenever the budget constraint binds. Failing to satisfy the Slutsky symmetry implies that they are not integrable in the sense of Hurwicz and Uzawa (1971), and thus, they are not rationalizable as a representative consumer’s behavior. The only possible economically and econometrically meaningful exception that we find is the simple multinomial logit model in which the mean utility is log-linear in prices.

Estimation of Random Coefficients Logit Demand Models with Interactive Fixed Effects

We extend the Berry, Levinsohn and Pakes (BLP, 1995) random coefficients discrete choice demand model, which underlies much recent empirical work in IO. We add interactive fixed effects in the form of a factor structure on the unobserved product characteristics. The interactive fixed effects can be arbitrarily correlated with the observed product characteristics (including price), which accommodates endogeneity and, at the same time, captures strong persistence in market shares across products and markets. We propose a two-step least squares-minimum distance (LS-MD) procedure to calculate the estimator. Our estimator is easy to compute, and Monte Carlo simulations show that it performs well. We consider an empirical illustration to US automobile demand.

Counterfactual Evaluation in Semiparametric Multinomial Choice Models

We propose using cyclic monotonicity, a convex-analytic property of the random utility choice model, to derive bounds on counterfactual choice probabilities in semiparametric multinomial choice models. These bounds are useful for typical counterfactual exercises in aggregate discrete-choice demand models. In our semiparametric approach, we do not specify the parametric distribution for the utility shocks, thus accommodating a wide variety of substitution patterns among alternatives. Computation of the counterfactual bounds is a tractable linear programming problem. We illustrate our approach in a series of Monte Carlo simulations and an empirical application using scanner data.

Industrial Reorganization: Learning About Patient Substitution Patterns from Natural Experiments

Despite their widespread usage, little is known about the predictive accuracy of different discrete choice demand models. To evaluate their performance, we use a series of natural disasters that unexpectedly removed hospitals from consumers' choice sets. We compare the model predictions of post-disaster behavior to the benchmark of actual post-disaster consumer behavior. Across our different settings, we find that models that allow for flexible interactions between patient characteristics and unobserved hospital quality perform the best and that it is important to use different classes of models. Further, the use of less accurate models could lead to more lax merger enforcement.