Randomized Algorithms for Lexicographic Inference

被引:4
作者
Kohli, Rajeev [1 ]
Boughanmi, Khaled [1 ]
Kohli, Vikram [2 ]
机构
[1] Columbia Univ, Grad Sch Business, New York, NY 10027 USA
[2] Northwestern Univ, McCormick Sch Engn, Dept Elect Engn & Comp Sci, Evanston, IL 60208 USA
关键词
lexicographic rules; noncompensatory choice; consumer search; discrete optimization; randomized algorithms; analysis of algorithms; maximum likelihood; MODELS; CHOICE; REPRESENTATION; SEARCH; JUDGMENT;
D O I
10.1287/opre.2018.1794
中图分类号
C93 [管理学];
学科分类号
12 ; 1201 ; 1202 ; 120202 ;
摘要
The inference of a lexicographic rule from paired comparisons, ranking, or choice data is a discrete optimization problem that generalizes the linear ordering problem. We develop an approach to its solution using randomized algorithms. First, we show that maximizing the expected value of a randomized solution is equivalent to solving the lexicographic inference problem. As a result, the discrete problem is transformed into a continuous and unconstrained nonlinear program that can be solved, possibly only to a local optimum, using nonlinear optimization methods. Second, we show that a maximum likelihood procedure, which runs in polynomial time, can be used to implement the randomized algorithm. The maximum likelihood value determines a lower bound on the performance ratio of the randomized algorithm. We employ the proposed approach to infer lexicographic rules for individuals using data from a choice experiment for electronic tablets. These rules obtain substantially better fit and predictions than a previously described greedy algorithm, a local search algorithm, and a multinomial logit model.
引用
收藏
页码:357 / 375
页数:19
相关论文
共 55 条
  • [1] Aggregating Inconsistent Information: Ranking and Clustering
    Ailon, Nir
    Charikar, Moses
    Newman, Alantha
    [J]. JOURNAL OF THE ACM, 2008, 55 (05)
  • [2] [Anonymous], 1954, DECISION PROCESSES
  • [3] [Anonymous], NONCON OPTIM ITS APP
  • [4] [Anonymous], 2004, J. Math. Model. Algorithms, DOI DOI 10.1023/B:JMMA.0000049426.06305.D8
  • [5] [Anonymous], 1998, Bayesian statistics
  • [6] VOTING SCHEMES FOR WHICH IT CAN BE DIFFICULT TO TELL WHO WON THE ELECTION
    BARTHOLDI, J
    TOVEY, CA
    TRICK, MA
    [J]. SOCIAL CHOICE AND WELFARE, 1989, 6 (02) : 157 - 165
  • [7] ASSESSING THE POTENTIAL DEMAND FOR ELECTRIC CARS
    BEGGS, S
    CARDELL, S
    HAUSMAN, J
    [J]. JOURNAL OF ECONOMETRICS, 1981, 17 (01) : 1 - 19
  • [8] Optimizing product line designs: Efficient methods and comparisons
    Belloni, Alexandre
    Freund, Robert
    Selove, Matthew
    Simester, Duncan
    [J]. MANAGEMENT SCIENCE, 2008, 54 (09) : 1544 - 1552
  • [9] Bertsimas D, 2019, OPER RES
  • [10] Robust Product Line Design
    Bertsimas, Dimitris
    Misic, Velibor V.
    [J]. OPERATIONS RESEARCH, 2017, 65 (01) : 19 - 37