Inference of Estimators Defined by Mathematical Programming

Abstract

We propose an inference procedure for estimators defined by mathematical programming problems, focusing on the important special cases of linear programming (LP) and quadratic programming (QP). In these settings, the coefficients in both the objective function and the constraints of the mathematical programming problem may be estimated from data and hence involve sampling error. Our inference approach exploits the characterization of the solutions to these programming problems by complementarity conditions; by doing so, we can transform the problem of doing inference on the solution of a constrained optimization problem (a non-standard inference problem) into one involving inference based on a set of inequalities with pre-estimated coefficients, which is much better understood. We evaluate the performance of our procedure in several Monte Carlo simulations and an empirical application to the classic portfolio selection problem in finance.

Do Lenders Still Discriminate? A Robust Approach for Assessing Differences in Menus

“…The derivation of complementary slackness conditions such as this can be found in standard texts on optimal transport/linear programming. In particular, Hsieh, Shi, and Shum (2017) use a similar set of conditions for their projection method. Based on this equivalency, the problem of finding critical values for the null hypothesis H 0 : ϕ ≤ ϕ 0 versus the alternative H a : ϕ > ϕ 0 can be written as the following LPCC:…”

Identification of the Average Treatment Effect When SUTVA Is Violated

“…We note that there are recent advances in statistical inference of partially identified parameters that deal with random linear programs of such form (Kaido et al 2019 4 or Hsieh et al 2018). Subsampling approaches may be used on the lower and upper bounds separately, as described in Lafférs (2019) or Demuynck (2015).…”

Seeing economic collapse and recovery through the lens of complexity economics

“…19 One can construct confidence set following Hsieh, Shi, and Shum (2018). In the interest of space, we do not pursue it in this paper.…”