Persistency Model and Its Applications in Choice Modeling

Abstract

Given a discrete maximization problem with a linear objective function where the coefficients are chosen randomly from a distribution, we would like to evaluate the expected optimal value and the marginal distribution of the optimal solution. We call this the persistency problem for a discrete optimization problem under uncertain objective, and the marginal probability mass function of the optimal solution is named the persistence value. In general, this is a difficult problem to solve, even if the distribution of the objective coefficient is well specified. In this paper, we solve a subclass of this problem when the distribution is assumed to belong to the class of distributions defined by given marginal distributions, or given marginal moment conditions. Under this model, we show that the persistency problem maximizing the expected objective value over the set of distributions can be solved via a concave maximization model. The persistency model solved using this formulation can be used to obtain important qualitative insights to the behavior of stochastic discrete optimization problems. We demonstrate how the approach can be used to obtain insights to problems in discrete choice modeling. Using a set of survey data from a transport choice modeling study, we calibrate the random utility model with choice probabilities obtained from the persistency model. Numerical results suggest that our persistency model is capable of obtaining estimates that perform as well, if not better, than classical methods, such as logit and cross-nested logit models. We can also use the persistency model to obtain choice probability estimates for more complex choice problems. We illustrate this on a stochastic knapsack problem, which is essentially a discrete choice problem under budget constraint.probability distribution, integer programming, utility preference, choice functions

Robustness to Dependency in Portfolio Optimization Using Overlapping Marginals

“…(a) The proof of Theorem 1 is inspired from the proofs in Bertsimas et al (2006) and Natarajan et al (2009b) for univariate marginals and Doan and Natarajan (2012) for nonoverlapping multivariate marginals. Theorem 1 extends these results to overlapping multivariate marginals.…”

Analysis of Discrete Choice Models: A Welfare-Based Framework

“…Recently, a new class of semi-parametric choice model (SCM) was proposed by Natarajan et al (2009). Unlike the random utility model where a certain distribution of the random utility ǫ is specified, in the semi-parametric choice model, one considers a set of distributions Θ for ǫ.…”

“…In the representative agent model, there is again a utility associated with each alternative, and the representative agent maximizes a weighted utility of the choice (which is a vector of proportions for each alternative) plus a regularization term, which typically encourages diversification of the choice (Anderson et al 1988). More recently, a class of semi-parametric models has been proposed (see Natarajan et al 2009). This model is similar to the random utility model.…”

A “Joint+Marginal” Approach to Parametric Polynomial Optimization

“…Then for instance, one might be interested to determine whether in an optimal solution x * (y) of P y , and for some index i ∈ I, one has x * i (y) = 1 (or x * i (y) = 0) for almost all values of the parameter y ∈ Y. In [3,17] the probability that x * k (y) is 1 is called the persistency of the boolean variable x * k (y) Corollary 3.7. Let K, Y be as in (2.2) and (3.1) respectively.…”

“…In particular, for discrete optimization problems where cost coefficients are random variables with joint distribution ϕ, some bounds on the expected optimal value have been obtained. More recently Natarajan et al [17] extended the earlier work in [3] to even 1991 Mathematics Subject Classification. 65 D15, 65 K05, 46 N10, 90 C22.…”