Julia codes for the paper "Data-Driven Sample Average Approximation with Covariate Information". Results in the paper were generated using Julia 0.6.4, JuMP 0.18.5, Gurobi 8.1.0, and GLMNet 0.3.0.
To generate results for the FI-SAA, N-SAA, ER-SAA, J-SAA & J+-SAA, or kNN-SAA methods, run the "main_fullinf_saa.jl", "main_nsaa.jl", "main_ersaa.jl", "main_jsaa.jl", or "main_knn.jl" file, respectively.
Parameter settings for each of the case studies are provided in the "params_case*" files within the Parameter settings folder, and the instances for each case study are provided in the "instances_case*" files within the Instances folder. To run each of the "main" files, set the variable "caseNum" to the appropriate value, and write a script to read the instances from the corresponding "instances" file.
COMING SOON: Script to read instances from the "instances_case*" files, and MATLAB codes for generating the figures in the paper from the generated result files.