|
on Discrete Choice Models |
| By: | Nikhil Agarwal; Pearl Z. Li; Paulo J. Somaini |
| Abstract: | Revealed preference arguments are commonly used when identifying models of both single-agent decisions and non-cooperative games. We develop general identification results for a large class of models that have a linearly separable payoff structure. Our model allows for both discrete and continuous choice sets. It incorporates widely studied models such as discrete and hedonic choice models, auctions, school choice mechanisms, oligopoly pricing and trading games. We characterize the identified set and show that point identification can be achieved either if the choice set is sufficiently rich or if a variable that shifts preferences is available. Our identification results also suggests an estimation approach. Finally, we implement this approach to estimate values in a combinatorial procurement auction for school lunches in Chile. |
| JEL: | C51 C57 |
| Date: | 2023–11 |
| URL: | http://d.repec.org/n?u=RePEc:nbr:nberwo:31868&r=dcm |
| By: | Kukushkin, Nikolai S. |
| Abstract: | Maximization of a preference relation on a given family of subsets of its domain defines a choice function. Assuming the domain to be a poset or a lattice, and considering subcomplete chains or sublattices as potential feasible sets, we study conditions ensuring the existence of optima, as well as properties of the choice function conducive to monotone comparative statics. Concerning optimization on chains, quite a number of characterization results are obtained; when it comes to lattices, we mostly obtain sufficient conditions. |
| Keywords: | preference relation; choice function; complete chain; complete lattice; quasisupermodularity; single crossing; monotone selection |
| JEL: | C61 C72 D11 |
| Date: | 2023–11–12 |
| URL: | http://d.repec.org/n?u=RePEc:pra:mprapa:119148&r=dcm |
| By: | Sid Kankanala |
| Abstract: | Latent variable models are widely used to account for unobserved determinants of economic behavior. Traditional nonparametric methods to estimate latent heterogeneity do not scale well into multidimensional settings. Distributional restrictions alleviate tractability concerns but may impart non-trivial misspecification bias. Motivated by these concerns, this paper introduces a quasi-Bayes approach to estimate a large class of multidimensional latent variable models. Our approach to quasi-Bayes is novel in that we center it around relating the characteristic function of observables to the distribution of unobservables. We propose a computationally attractive class of priors that are supported on Gaussian mixtures and derive contraction rates for a variety of latent variable models. |
| Date: | 2023–11 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2311.06831&r=dcm |
| By: | Kaiwen Hou |
| Abstract: | This manuscript presents an advanced framework for Bayesian learning by incorporating action and state-dependent signal variances into decision-making models. This framework is pivotal in understanding complex data-feedback loops and decision-making processes in various economic systems. Through a series of examples, we demonstrate the versatility of this approach in different contexts, ranging from simple Bayesian updating in stable environments to complex models involving social learning and state-dependent uncertainties. The paper uniquely contributes to the understanding of the nuanced interplay between data, actions, outcomes, and the inherent uncertainty in economic models. |
| Date: | 2023–11 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2311.12878&r=dcm |