PUBLICATIONS

Preprint

1. Variable selection in convex nonparametric least squares via structured Lasso penalty: An application to the Swedish electricity market

   Zhiqiang Liao

   2024

   arXiv
2. Overfitting Reduction in Convex Regression

   Zhiqiang Liao, Sheng Dai, Lim Eunji, and 1 more author

   2024

   arXiv

2024

1. Convex support vector regression

   Zhiqiang Liao, Sheng Dai, and Timo Kuosmanen

   European Journal of Operational Research 2024

   Abs HTML

   Nonparametric regression subject to convexity or concavity constraints is increasingly popular in economics, finance, operations research, machine learning, and statistics. However, the conventional convex regression based on the least squares loss function often suffers from overfitting and outliers. This paper proposes to address these two issues by introducing the convex support vector regression (CSVR) method, which effectively combines the key elements of convex regression and support vector regression. Numerical experiments demonstrate the performance of CSVR in prediction accuracy and robustness that compares favorably with other state-of-the-art methods.

2023

1. A contextual Choquet integral-based preference learning model considering both criteria interactions and the compromise effects of decision-makers

   Zhiqiang Liao, Huchang Liao, and Xinli Zhang

   Expert Systems with Applications 2023

   Abs HTML PDF

   Preference learning has been widely employed to predict decision-makers’ preferences from historical information. This study develops a preference learning model for multiple criteria decision analysis where the decision-maker is supposed to be bounded rational and criteria are not completely independent of each other. The contextual Choquet integral is used as the aggregation function to address criteria interactions. The robust-ordinal-regression (ROR) technique is then applied to learn the preferences of decision-makers from the given preference data and provide robust decision recommendations. The proposed approach is illustrated by a numerical study concerning sustainable product evaluation.

2022

1. Compromise solutions for stochastic multicriteria acceptability analysis with uncertain preferences and nonmonotonic criteria

   Zhiqiang Liao, Huchang Liao, and Benjamin Lev

   International Transactions in Operational Research 2022

   Abs HTML

   As one of the most popular multiple criteria decision-making (MCDM) methods, the stochastic multicriteria acceptability analysis (SMAA) is powerful in dealing with the uncertain preference information of decision makers. However, how to find compromise solutions for MCDM problems remains an open question in the SMAA family. A compromise solution refers to an alternative that is the closest to the ideal solution. In this paper, we propose an improved SMAA method, called the SMAA with compromise solutions (SMAA-CO), to determine compromise solutions for MCDM problems with the uncertain preferences of decision makers and conflicting criteria. The SMAA-CO uses nonmonotonic additive value functions to depict decision makers’ ideal values for reference points. We apply the proposed method to capture multiple decision makers’ uncertain preference information. An illustrative example concerning supplier selection in the pulp and paper industry is used to show the applicability of the proposed method. Some managerial insights are given.

2020

1. A Choquet integral-based hesitant fuzzy gained and lost dominance score method for multi-criteria group decision making considering the risk preferences of experts: Case study of higher business education evaluation

   Zhiqiang Liao, Huchang Liao, Ming Tang, and 2 more authors

   Information Fusion 2020

   Abs HTML

   With the rapid development of higher business education, higher business education evaluation has attracted considerable attention of researchers and practitioners. The higher business education evaluation is an essential part of the development of a business school, which has a direct impact on its resource distribution. The higher business education evaluation can be considered as a multiple criteria group decision making (MCGDM) problem that involves a group of experts. Due to the complexity of the decision-making problem, decision criteria are not fully independent to each other, and the assumption of complete rationality of experts is usually invalid in many situations. In this paper, we propose a Choquet integral-based hesitant fuzzy gained and lost dominance score method to address the two important issues regarding the interactions among criteria and the behavior preference characteristics of experts in MCGDM problems. Firstly, a comprehensive distance measure of hesitant fuzzy sets is introduced by considering the relative importance of two separations. Then, a Choquet integral-based hesitant fuzzy gained and lost dominance score method based on the prospect theory is proposed to address the MCGDM problems in which experts make decision with the risk preference psychology. Finally, an illustrative example of higher business education evaluation is provided to demonstrate the applicability of the proposed method, and the sensitivity and comparative analysis are also completed to verify the validity of the proposed method.

2019

1. A HESITANT FUZZY LINGUISTIC CHOQUET INTEGRAL-BASED MULTIMOORA METHOD FOR MULTIPLE CRITERIA DECISION MAKING AND ITS APPLICATION IN TALENT SELECTION.

   Zhiqiang Liao, Huchang Liao, Xunjie Gou, and 2 more authors

   Economic Computation & Economic Cybernetics Studies & Research 2019

   Abs HTML PDF

   In qualitative multiple criteria decision making (MCDM) process, dealing with complex linguistic evaluations and modeling interaction phenomena among criteria are two important issues. The purpose of this study is to introduce two Choquet integral operators for HFLTSs to tackle interactions among criteria and then implement these operators to a well-known MCDM method, namely, the MULTIMOORA (multi-objective optimization by ratio analysis plus the full multiplicative form). Firstly, we introduce new operations for HFLTSs based on some linguistic scale functions. Then, we define two hesitant fuzzy linguistic Choquet integral operators and propose a novel score function for hesitant fuzzy linguistic elements. Afterwards, we improve the HFL-MULTIMOORA method to handle the hesitant fuzzy linguistic MCDM problems in which the criteria are inter-dependent. Finally, an illustration regarding the human resource development at Sichuan University is shown to verify the applicability and validity of the proposed method.