Journal of Choice Modelling最新文献_第3页

Understanding the decision-making process of choice modellers 理解选择建模者的决策过程

IF 2.4 3区经济学 Q1 ECONOMICS

Journal of Choice Modelling

Pub Date : 2025-09-01 Epub Date: 2025-08-25 DOI: 10.1016/j.jocm.2025.100562

Gabriel Nova , Sander van Cranenburgh , Stephane Hess

Choice Modelling is a widely used framework for understanding human choice behaviour across disciplines. Building a choice model is a complex, semi-structured process that involves a combination of prior assumptions, behavioural theories, and statistical methods. This complex set of decisions, coupled with diverse workflows, can lead to substantial variability in model outcomes. To investigate these modelling processes, we introduce the Discrete Choice Modelling Serious Game (DCM-SG), a novel tool that mimics the workflow of choice modellers and tracks the modelling decisions participants make. In our application, participants developed models to estimate willingness-to-pay values for reducing noise pollution. Their actions were tracked, enabling analysis of workflow patterns and modelling strategies. Forty participants, most with over five years of experience, completed the game. Our contributions are twofold. Methodologically, the DCM-SG captures sequential data on modellers’ workflows, which we analyse using telemetry and sequential pattern mining techniques to uncover dynamic patterns of in-game tool usage, phase transitions, and model specification approaches. Substantively, there was a strong preference for data visualisation and frequent specification of simpler models (Multinomial Logit), alongside attempts to specify more complex models. These findings suggest that in time-constrained or resource-limited settings, modellers may underexplore important factors such as covariates, non-linearities, and complex specifications. Moreover, participants who engaged more thoroughly in data exploration and iterative model comparison consistently achieved superior model fit and parsimony. These results demonstrate how sequential data from the DCM-SG can uncover variations in modelling practices and provide a foundation for understanding the art of choice modelling.

选择模型是一个广泛使用的框架，用于理解跨学科的人类选择行为。建立选择模型是一个复杂的、半结构化的过程，涉及到先验假设、行为理论和统计方法的结合。这种复杂的决策集，加上不同的工作流程，会导致模型结果的大量变化。为了研究这些建模过程，我们引入了离散选择建模严肃博弈（DCM-SG），这是一种模仿选择建模者工作流程并跟踪参与者建模决策的新工具。在我们的应用中，参与者开发了模型来估计减少噪音污染的支付意愿值。他们的行为被跟踪，从而能够分析工作流程模式和建模策略。40名参与者完成了这款游戏，其中大多数都有5年以上的经验。我们的贡献是双重的。在方法上，DCM-SG捕获建模人员工作流程上的顺序数据，我们使用遥测和顺序模式挖掘技术对其进行分析，以揭示游戏内工具使用、相变和模型规范方法的动态模式。从本质上讲，人们更倾向于数据可视化和简单模型（多项式Logit）的频繁规范，同时试图指定更复杂的模型。这些发现表明，在时间有限或资源有限的情况下，建模者可能没有充分考虑协变量、非线性和复杂规格等重要因素。此外，更深入地参与数据探索和迭代模型比较的参与者一致地获得了更好的模型拟合和简约性。这些结果展示了DCM-SG的顺序数据如何揭示建模实践中的变化，并为理解选择建模的艺术提供了基础。

{"title":"Understanding the decision-making process of choice modellers","authors":"Gabriel Nova , Sander van Cranenburgh , Stephane Hess","doi":"10.1016/j.jocm.2025.100562","DOIUrl":"10.1016/j.jocm.2025.100562","url":null,"abstract":"<div><div>Choice Modelling is a widely used framework for understanding human choice behaviour across disciplines. Building a choice model is a complex, semi-structured process that involves a combination of prior assumptions, behavioural theories, and statistical methods. This complex set of decisions, coupled with diverse workflows, can lead to substantial variability in model outcomes. To investigate these modelling processes, we introduce the Discrete Choice Modelling Serious Game (DCM-SG), a novel tool that mimics the workflow of choice modellers and tracks the modelling decisions participants make. In our application, participants developed models to estimate willingness-to-pay values for reducing noise pollution. Their actions were tracked, enabling analysis of workflow patterns and modelling strategies. Forty participants, most with over five years of experience, completed the game. Our contributions are twofold. Methodologically, the DCM-SG captures sequential data on modellers’ workflows, which we analyse using telemetry and sequential pattern mining techniques to uncover dynamic patterns of in-game tool usage, phase transitions, and model specification approaches. Substantively, there was a strong preference for data visualisation and frequent specification of simpler models (Multinomial Logit), alongside attempts to specify more complex models. These findings suggest that in time-constrained or resource-limited settings, modellers may underexplore important factors such as covariates, non-linearities, and complex specifications. Moreover, participants who engaged more thoroughly in data exploration and iterative model comparison consistently achieved superior model fit and parsimony. These results demonstrate how sequential data from the DCM-SG can uncover variations in modelling practices and provide a foundation for understanding the art of choice modelling.</div></div>","PeriodicalId":46863,"journal":{"name":"Journal of Choice Modelling","volume":"56 ","pages":"Article 100562"},"PeriodicalIF":2.4,"publicationDate":"2025-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144893393","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Exploring the choice landscape: Anchoring and framing effects on search behavior in complex choices 探索选择景观：复杂选择中搜索行为的锚定和框架效应

IF 2.8 3区经济学 Q1 ECONOMICS

Journal of Choice Modelling

Pub Date : 2025-06-01 Epub Date: 2025-03-29 DOI: 10.1016/j.jocm.2025.100549

Nikolos Gurney , John H. Miller , David V. Pynadath

Complex choices entail consideration of multiple variables characterized by non-linear interactions. Prior research into such choices neglects two categories of major choice determinants: heuristics and biases. Our research integrates such behavioral determinants of decision making into the study of complex choices. We approach this challenge using a new experimental paradigm in which participants tune on-screen dials representing choice variables. Through their tuning efforts, participants can discover information about the choice space. Specifically, this paradigm allows us to document how anchoring and framing effects impact complex choice processes. We find that loss framing incites participants to expend more effort exploring choice variables before selecting an option. An aspirational anchor, on the other hand, correlates with decreased search effort. Additionally, when participants did not have an aspirational anchor, the data suggest they anchor on prior experience.

复杂的选择需要考虑以非线性相互作用为特征的多个变量。先前对这类选择的研究忽略了两类主要的选择决定因素：启发式和偏见。我们的研究将这些决策的行为决定因素整合到复杂选择的研究中。我们使用一种新的实验范式来应对这一挑战，在这种范式中，参与者调整代表选择变量的屏幕刻度盘。通过调优工作，参与者可以发现有关选择空间的信息。具体来说，这个范例允许我们记录锚定和框架效应如何影响复杂的选择过程。我们发现损失框架激励参与者在选择一个选项之前花费更多的精力来探索选择变量。另一方面，一个有抱负的锚与减少搜索努力相关。此外，当参与者没有理想锚时，数据表明他们会根据先前的经验进行锚定。

引用次数: 0

Cost vector effects in forced-choice discrete choice experiments: Assessing the acceptability of future glyphosate policies 强迫选择离散选择实验中的成本向量效应：评估未来草甘膦政策的可接受性

IF 2.8 3区经济学 Q1 ECONOMICS

Journal of Choice Modelling

Pub Date : 2025-06-01 Epub Date: 2025-04-11 DOI: 10.1016/j.jocm.2025.100550

Vincent Martinet , Maïa David , Vincent Mermet-Bijon , Romain Crastes Dit Sourd

One way to evaluate future policies that significantly deviate from the status quo is through discrete choice experiments (DCEs) with a reference policy featuring a positive cost and no opt-out option. This study examines how the design of the cost vector, particularly the cost assigned to the reference policy, influences DCE outcomes in this context. Focusing on glyphosate phase-out policies in France, we compare a strict ban (used as the reference policy) with taxation alternatives. Using a split-sample design with two groups of 500 individuals, we analyze how variations in the ban’s cost and the associated cost range affect welfare estimates. Our findings reveal that while overall preference rankings remain consistent across samples, willingness-to-pay for some attributes increases when the reference policy’s cost rises. We explore potential drivers of this effect, including the inability to choke off demand for the ban, strategic biases, attribute non-attendance, relative evaluation, and anchoring bias. The results suggest that relative evaluation and anchoring bias are the most likely explanations for the observed differences. These findings provide methodological insights for addressing cost vector effects in DCEs.

评估明显偏离现状的未来政策的一种方法是通过离散选择实验（DCEs），其中包含一个具有正成本且无选择退出选项的参考政策。本研究探讨了成本向量的设计，特别是分配给参考政策的成本，在这种情况下如何影响DCE结果。着眼于法国的草甘膦淘汰政策，我们将严格的禁令（用作参考政策）与税收替代方案进行了比较。采用两组500人的分离样本设计，我们分析了禁令成本和相关成本范围的变化如何影响福利估计。我们的研究结果表明，虽然样本的总体偏好排名保持一致，但当参考政策的成本上升时，某些属性的支付意愿会增加。我们探讨了这种影响的潜在驱动因素，包括无法抑制对禁令的需求、战略偏见、属性不出席、相对评估和锚定偏见。结果表明，相对评价和锚定偏差是对观察到的差异最可能的解释。这些发现为解决dce中的成本向量效应提供了方法论见解。

{"title":"Cost vector effects in forced-choice discrete choice experiments: Assessing the acceptability of future glyphosate policies","authors":"Vincent Martinet , Maïa David , Vincent Mermet-Bijon , Romain Crastes Dit Sourd","doi":"10.1016/j.jocm.2025.100550","DOIUrl":"10.1016/j.jocm.2025.100550","url":null,"abstract":"<div><div>One way to evaluate future policies that significantly deviate from the status quo is through discrete choice experiments (DCEs) with a reference policy featuring a positive cost and no opt-out option. This study examines how the design of the cost vector, particularly the cost assigned to the reference policy, influences DCE outcomes in this context. Focusing on glyphosate phase-out policies in France, we compare a strict ban (used as the reference policy) with taxation alternatives. Using a split-sample design with two groups of 500 individuals, we analyze how variations in the ban’s cost and the associated cost range affect welfare estimates. Our findings reveal that while overall preference rankings remain consistent across samples, willingness-to-pay for some attributes increases when the reference policy’s cost rises. We explore potential drivers of this effect, including the inability to choke off demand for the ban, strategic biases, attribute non-attendance, relative evaluation, and anchoring bias. The results suggest that relative evaluation and anchoring bias are the most likely explanations for the observed differences. These findings provide methodological insights for addressing cost vector effects in DCEs.</div></div>","PeriodicalId":46863,"journal":{"name":"Journal of Choice Modelling","volume":"55 ","pages":"Article 100550"},"PeriodicalIF":2.8,"publicationDate":"2025-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143816900","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

An Experience-Based Choice Model (EBCM): Formulation, identification, behavioural insights and well-being assessment 基于经验的选择模型（EBCM）：制定，识别，行为洞察和福祉评估

IF 2.8 3区经济学 Q1 ECONOMICS

Journal of Choice Modelling

Pub Date : 2025-06-01 Epub Date: 2025-04-19 DOI: 10.1016/j.jocm.2025.100552

Bastian Henriquez-Jara , C. Angelo Guevara

The Experience-Based Choice Model (EBCM) accounts for the underlying learning process within choices, where the main learning unit is the instant utility: a representation of the hedonic momentary feelings. It merges Kahneman’s (Kahneman et al., 1997) notions of experienced utility with the Integrated Choice and Latent Variables (ICLV) framework, drawing on the current knowledge about the use of physiological signals to measure psychological states. EBCM allows us to model choices as a function of the learnt descriptive information of attributes and the hedonic experiences from past choices and to obtain a measure of the experience which can be used to analyse Subjective Well-Being (SWB). In this article, we first implement the model as a Markovian learning process in which the individual uses the remembered utility to update the decision utility at a certain learning rate while aggregating the instant utilities using an Instance-Based Learning (IBL) approach. We discuss specification and identification issues with Monte Carlo simulations. Besides, we use numerical experiments to show how this framework allows us to capture challenging behavioural phenomena such as the duality of habitual and goal-directed behaviour, the hot stove effect, the status quo bias, and the recency effect. Then, we discuss how this formulation contributes to the analysis of SWB. EBCM can help to assess SWB, particularly in situations where traditional assumptions of consumer rationality may not hold and when consumers are captive. Finally, new lines of research are discussed.

基于经验的选择模型（EBCM）描述了选择过程中的基本学习过程，其中主要的学习单位是即时效用：一种享乐主义瞬间感受的代表。它融合了卡尼曼（Kahneman 等人，1997 年）的经验效用概念和综合选择与潜变量（ICLV）框架，并借鉴了当前使用生理信号测量心理状态的知识。EBCM 使我们能够将选择建模为所学属性描述信息和过去选择的享乐体验的函数，并获得可用于分析主观幸福感 (SWB) 的体验测量值。在本文中，我们首先将该模型作为一个马尔可夫学习过程来实现，在该过程中，个体使用记忆效用以一定的学习率更新决策效用，同时使用基于实例的学习（IBL）方法汇总即时效用。我们通过蒙特卡罗模拟讨论了规范和识别问题。此外，我们还利用数值实验来说明这一框架如何让我们捕捉到具有挑战性的行为现象，如习惯行为和目标导向行为的二元性、热炉效应、现状偏差和追忆效应。然后，我们将讨论这一表述如何有助于分析 SWB。EBCM 有助于评估 SWB，尤其是在消费者理性的传统假设可能不成立的情况下和消费者被俘虏的情况下。最后，我们将讨论新的研究方向。

{"title":"An Experience-Based Choice Model (EBCM): Formulation, identification, behavioural insights and well-being assessment","authors":"Bastian Henriquez-Jara , C. Angelo Guevara","doi":"10.1016/j.jocm.2025.100552","DOIUrl":"10.1016/j.jocm.2025.100552","url":null,"abstract":"<div><div>The Experience-Based Choice Model (EBCM) accounts for the underlying learning process within choices, where the main learning unit is the instant utility: a representation of the hedonic momentary feelings. It merges Kahneman’s (Kahneman et al., 1997) notions of experienced utility with the Integrated Choice and Latent Variables (ICLV) framework, drawing on the current knowledge about the use of physiological signals to measure psychological states. EBCM allows us to model choices as a function of the learnt descriptive information of attributes and the hedonic experiences from past choices and to obtain a measure of the experience which can be used to analyse Subjective Well-Being (SWB). In this article, we first implement the model as a Markovian learning process in which the individual uses the remembered utility to update the decision utility at a certain learning rate while aggregating the instant utilities using an Instance-Based Learning (IBL) approach. We discuss specification and identification issues with Monte Carlo simulations. Besides, we use numerical experiments to show how this framework allows us to capture challenging behavioural phenomena such as the duality of habitual and goal-directed behaviour, the hot stove effect, the status quo bias, and the recency effect. Then, we discuss how this formulation contributes to the analysis of SWB. EBCM can help to assess SWB, particularly in situations where traditional assumptions of consumer rationality may not hold and when consumers are captive. Finally, new lines of research are discussed.</div></div>","PeriodicalId":46863,"journal":{"name":"Journal of Choice Modelling","volume":"55 ","pages":"Article 100552"},"PeriodicalIF":2.8,"publicationDate":"2025-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143850539","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Modelling viewing patterns of serial TV dramas considering live viewing and time shifting 考虑直播和时移的电视剧观看模式建模

IF 2.8 3区经济学 Q1 ECONOMICS

Journal of Choice Modelling

Pub Date : 2025-06-01 Epub Date: 2025-03-19 DOI: 10.1016/j.jocm.2025.100548

Ryosuke Igari

In this study, we focus on the segmentation of viewing behavior for serial TV dramas at the individual level, considering both live and time-shift viewing. In particular, many consumers have recently engaged in time shifting after a program is broadcast, in addition to live viewing. Time-shift viewing is prominent in drama viewing because drama viewing may be less urgent. However, few studies that focus on individual-level viewing choice models have considered time shifting and no studies have utilized segmentation to analyze heterogeneous viewing patterns that consider live viewing and time-shifts. We use a multinomial probit (MNP) model to capture viewing or not viewing and viewing method (live or time shifting). Furthermore, a latent growth curve model and a latent class model are included in the MNP model to represent heterogeneous patterns of viewer retention and dropout as episodes progressed. We analyze six individual dramas and a simultaneous analysis of multiple dramas broadcast in 2019. The results show that segments are affected by popularity and growth curve models, as well as differences in viewing rate trends by episode and segment. The findings provide information that can support marketing research and media strategies.

在本研究中，我们重点从个体层面对连续剧的收视行为进行了细分，同时考虑了直播收视和时移收视。特别是，最近许多消费者在节目播出后，除了直播观看外，还进行了时移观看。时移收视在电视剧收视中比较突出，因为电视剧收视可能不那么紧迫。然而，关注个体层面收视选择模型的研究很少考虑时移问题，也没有研究利用细分来分析考虑直播收视和时移的异质收视模式。我们使用多叉概率（MNP）模型来捕捉观看与否和观看方式（直播或时移）。此外，在 MNP 模型中还加入了潜增长曲线模型和潜类模型，以表示随着剧集的播出，观众保留和退出的异质性模式。我们分析了六部单独的电视剧，并对 2019 年播出的多部电视剧进行了同步分析。结果显示，细分市场受到人气和增长曲线模型的影响，不同剧集和细分市场的收视率趋势也存在差异。研究结果提供的信息可为营销研究和媒体战略提供支持。

{"title":"Modelling viewing patterns of serial TV dramas considering live viewing and time shifting","authors":"Ryosuke Igari","doi":"10.1016/j.jocm.2025.100548","DOIUrl":"10.1016/j.jocm.2025.100548","url":null,"abstract":"<div><div>In this study, we focus on the segmentation of viewing behavior for serial TV dramas at the individual level, considering both live and time-shift viewing. In particular, many consumers have recently engaged in time shifting after a program is broadcast, in addition to live viewing. Time-shift viewing is prominent in drama viewing because drama viewing may be less urgent. However, few studies that focus on individual-level viewing choice models have considered time shifting and no studies have utilized segmentation to analyze heterogeneous viewing patterns that consider live viewing and time-shifts. We use a multinomial probit (MNP) model to capture viewing or not viewing and viewing method (live or time shifting). Furthermore, a latent growth curve model and a latent class model are included in the MNP model to represent heterogeneous patterns of viewer retention and dropout as episodes progressed. We analyze six individual dramas and a simultaneous analysis of multiple dramas broadcast in 2019. The results show that segments are affected by popularity and growth curve models, as well as differences in viewing rate trends by episode and segment. The findings provide information that can support marketing research and media strategies.</div></div>","PeriodicalId":46863,"journal":{"name":"Journal of Choice Modelling","volume":"55 ","pages":"Article 100548"},"PeriodicalIF":2.8,"publicationDate":"2025-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143687087","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Constructing Bayesian optimal designs for discrete choice experiments by simulated annealing 用模拟退火构造离散选择实验的贝叶斯最优设计

IF 2.8 3区经济学 Q1 ECONOMICS

Journal of Choice Modelling

Pub Date : 2025-06-01 Epub Date: 2025-04-04 DOI: 10.1016/j.jocm.2025.100551

Yicheng Mao , Roselinde Kessels , Tom van der Zanden

Discrete choice experiments (DCEs) investigate the attributes that influence individuals’ choices when selecting among various options. To enhance the quality of the estimated choice models, researchers opt for Bayesian optimal designs that utilize existing information about the attributes’ preferences. Given the nonlinear nature of choice models, the construction of an appropriate design requires efficient algorithms. Among these, the coordinate-exchange (CE) algorithm is commonly employed for constructing designs based on the MNL model. However, as a hill-climbing method, the CE algorithm tends to quickly converge to local optima, potentially limiting the quality of the resulting designs. We propose the use of a simulated annealing (SA) algorithm to construct Bayesian optimal designs. This algorithm accepts both superior and inferior solutions, avoiding premature convergence and allowing a more thorough exploration of potential solutions. Consequently, it ultimately obtains higher-quality choice designs compared to the CE algorithm. Our work represents the first application of an SA algorithm in constructing Bayesian optimal designs for DCEs. Through extensive computational experiments, we demonstrate that the SA designs generally outperform the CE designs in terms of statistical efficiency, especially when the prior preference information is highly uncertain.

离散选择实验（DCE）研究的是影响个人在各种选项中做出选择的属性。为了提高估计选择模型的质量，研究人员会选择贝叶斯优化设计，利用现有的属性偏好信息。鉴于选择模型的非线性特点，构建合适的设计需要高效的算法。其中，坐标交换（CE）算法通常用于构建基于 MNL 模型的设计。然而，作为一种爬山方法，CE 算法往往会迅速收敛到局部最优，从而可能限制设计结果的质量。我们建议使用模拟退火（SA）算法来构建贝叶斯最优设计。该算法同时接受优劣两种解决方案，避免了过早收敛，并允许对潜在解决方案进行更彻底的探索。因此，与 CE 算法相比，它最终能获得更高质量的选择设计。我们的工作代表了 SA 算法在构建 DCE 的贝叶斯最优设计中的首次应用。通过大量的计算实验，我们证明了 SA 设计在统计效率方面普遍优于 CE 设计，尤其是在先验偏好信息高度不确定的情况下。

{"title":"Constructing Bayesian optimal designs for discrete choice experiments by simulated annealing","authors":"Yicheng Mao , Roselinde Kessels , Tom van der Zanden","doi":"10.1016/j.jocm.2025.100551","DOIUrl":"10.1016/j.jocm.2025.100551","url":null,"abstract":"<div><div>Discrete choice experiments (DCEs) investigate the attributes that influence individuals’ choices when selecting among various options. To enhance the quality of the estimated choice models, researchers opt for Bayesian optimal designs that utilize existing information about the attributes’ preferences. Given the nonlinear nature of choice models, the construction of an appropriate design requires efficient algorithms. Among these, the coordinate-exchange (CE) algorithm is commonly employed for constructing designs based on the MNL model. However, as a hill-climbing method, the CE algorithm tends to quickly converge to local optima, potentially limiting the quality of the resulting designs. We propose the use of a simulated annealing (SA) algorithm to construct Bayesian optimal designs. This algorithm accepts both superior and inferior solutions, avoiding premature convergence and allowing a more thorough exploration of potential solutions. Consequently, it ultimately obtains higher-quality choice designs compared to the CE algorithm. Our work represents the first application of an SA algorithm in constructing Bayesian optimal designs for DCEs. Through extensive computational experiments, we demonstrate that the SA designs generally outperform the CE designs in terms of statistical efficiency, especially when the prior preference information is highly uncertain.</div></div>","PeriodicalId":46863,"journal":{"name":"Journal of Choice Modelling","volume":"55 ","pages":"Article 100551"},"PeriodicalIF":2.8,"publicationDate":"2025-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143768976","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Social finance in emerging markets: Insights into Chinese individual investor preferences with broader implications 新兴市场的社会融资：对中国个人投资者偏好的洞察与更广泛的影响

IF 2.8 3区经济学 Q1 ECONOMICS

Journal of Choice Modelling

Pub Date : 2025-03-01 Epub Date: 2024-12-05 DOI: 10.1016/j.jocm.2024.100534

Yaoming Liang , Ruiqi Chen , Senbin Zhang , Hongfu Liu , Li Han

Social finance, which integrates social impact with financial returns, is increasingly recognized as a catalyst for sustainable socio-economic development. This study presents an empirical analysis of Chinese individual investors' preferences for social finance products, using a customized choice experiment. The results reveal a strong investor preference for products with high annualized returns, short investment tenures, a focus on the healthcare sector, government backing, and flexible redemption options. By applying latent class analysis, we identify three distinct investor segments: Flexible Savers, Holistic Benefit Assessors, and Economic Return Seekers. The study also examines how demographic characteristics such as age, education, income, risk tolerance, philanthropic involvement, and awareness of social finance are associated with distinct investor segments. The findings suggest that future efforts should focus on designing social finance products that align with the diverse needs of these investor segments, thereby enhancing the appeal and effectiveness of social finance initiatives. Such tailored strategies could play a pivotal role in mobilizing private capital for social finance, maximizing its potential to drive sustainable socio-economic progress in emerging markets.

社会融资将社会影响与财务回报结合起来，越来越被认为是可持续社会经济发展的催化剂。本研究采用定制选择实验对中国个人投资者对社会金融产品的偏好进行了实证分析。结果显示，投资者强烈偏好具有高年化回报、投资期限短、关注医疗保健行业、政府支持和灵活赎回选择的产品。通过应用潜在类别分析，我们确定了三个不同的投资者群体：灵活储蓄者，整体利益评估者和经济回报寻求者。该研究还考察了年龄、教育程度、收入、风险承受能力、慈善参与和社会金融意识等人口特征如何与不同的投资者群体相关联。研究结果表明，未来的工作应侧重于设计符合这些投资者群体多样化需求的社会金融产品，从而提高社会金融举措的吸引力和有效性。这种量身定制的战略可以在动员私人资本进行社会融资方面发挥关键作用，最大限度地发挥其推动新兴市场可持续社会经济进步的潜力。

{"title":"Social finance in emerging markets: Insights into Chinese individual investor preferences with broader implications","authors":"Yaoming Liang , Ruiqi Chen , Senbin Zhang , Hongfu Liu , Li Han","doi":"10.1016/j.jocm.2024.100534","DOIUrl":"10.1016/j.jocm.2024.100534","url":null,"abstract":"<div><div>Social finance, which integrates social impact with financial returns, is increasingly recognized as a catalyst for sustainable socio-economic development. This study presents an empirical analysis of Chinese individual investors' preferences for social finance products, using a customized choice experiment. The results reveal a strong investor preference for products with high annualized returns, short investment tenures, a focus on the healthcare sector, government backing, and flexible redemption options. By applying latent class analysis, we identify three distinct investor segments: Flexible Savers, Holistic Benefit Assessors, and Economic Return Seekers. The study also examines how demographic characteristics such as age, education, income, risk tolerance, philanthropic involvement, and awareness of social finance are associated with distinct investor segments. The findings suggest that future efforts should focus on designing social finance products that align with the diverse needs of these investor segments, thereby enhancing the appeal and effectiveness of social finance initiatives. Such tailored strategies could play a pivotal role in mobilizing private capital for social finance, maximizing its potential to drive sustainable socio-economic progress in emerging markets.</div></div>","PeriodicalId":46863,"journal":{"name":"Journal of Choice Modelling","volume":"54 ","pages":"Article 100534"},"PeriodicalIF":2.8,"publicationDate":"2025-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143146673","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Context-aware Bayesian mixed multinomial logit model 上下文感知贝叶斯混合多项逻辑模型

IF 2.8 3区经济学 Q1 ECONOMICS

Journal of Choice Modelling

Pub Date : 2025-03-01 Epub Date: 2024-12-12 DOI: 10.1016/j.jocm.2024.100536

Mirosława Łukawska, Anders Fjendbo Jensen, Filipe Rodrigues

Traditional choice models often entail the assumption that the preference parameters of the decision-maker are constant throughout time and across different choice situations, which may be too strong for certain choice modelling applications. This paper proposes an effective approach to model systematic, context-dependent heterogeneity, thereby introducing the concept of the context-aware Bayesian mixed multinomial logit model (C-MMNL). In this model, a neural network maps contextual information to interpretable shifts in the preference parameters of each individual in each choice occasion. The proposed model offers several key advantages. First, it supports both continuous and discrete variables, as well as complex non-linear interactions between both types of variables. Secondly, each context specification is considered jointly as a whole by the neural network, rather than each variable being considered independently. Finally, since the neural network parameters are shared across all decision-makers, it can leverage information from other decision-makers to infer the effect of a particular context on a particular decision-maker. Even though the context-aware Bayesian mixed multinomial logit model allows for flexible interactions between attributes, the increase in computational complexity is minor, compared to the mixed multinomial logit model. We illustrate the concept and interpretation of the proposed model in a simulation study. We furthermore present a real-world case study from the travel behaviour domain — a bicycle route choice model, based on a large-scale, crowdsourced dataset of GPS trajectories including 119,448 trips made by 8555 cyclists.

传统的选择模型通常需要假设决策者的偏好参数在整个时间和不同的选择情况下是恒定的，这对于某些选择建模应用可能太强了。本文提出了一种有效的方法来模拟系统的、上下文相关的异质性，从而引入了上下文感知贝叶斯混合多项逻辑模型（C-MMNL）的概念。在这个模型中，神经网络将上下文信息映射到每个个体在每个选择场合的偏好参数的可解释变化。提出的模型提供了几个关键的优点。首先，它支持连续变量和离散变量，以及两种变量之间复杂的非线性相互作用。其次，神经网络将每个上下文规范作为一个整体来考虑，而不是单独考虑每个变量。最后，由于神经网络参数在所有决策者之间共享，它可以利用来自其他决策者的信息来推断特定上下文对特定决策者的影响。尽管上下文感知的Bayesian混合多项logit模型允许属性之间灵活的交互，但与混合多项logit模型相比，计算复杂性的增加很小。我们在模拟研究中说明了所提出模型的概念和解释。我们进一步提出了一个来自旅行行为领域的现实案例研究——一个自行车路线选择模型，该模型基于一个大规模的、众包的GPS轨迹数据集，包括8555名骑自行车的人的119,448次旅行。

{"title":"Context-aware Bayesian mixed multinomial logit model","authors":"Mirosława Łukawska, Anders Fjendbo Jensen, Filipe Rodrigues","doi":"10.1016/j.jocm.2024.100536","DOIUrl":"10.1016/j.jocm.2024.100536","url":null,"abstract":"<div><div>Traditional choice models often entail the assumption that the preference parameters of the decision-maker are constant throughout time and across different choice situations, which may be too strong for certain choice modelling applications. This paper proposes an effective approach to model systematic, context-dependent heterogeneity, thereby introducing the concept of the context-aware Bayesian mixed multinomial logit model (C-MMNL). In this model, a neural network maps contextual information to interpretable shifts in the preference parameters of each individual in each choice occasion. The proposed model offers several key advantages. First, it supports both continuous and discrete variables, as well as complex non-linear interactions between both types of variables. Secondly, each context specification is considered jointly as a whole by the neural network, rather than each variable being considered independently. Finally, since the neural network parameters are shared across all decision-makers, it can leverage information from other decision-makers to infer the effect of a particular context on a particular decision-maker. Even though the context-aware Bayesian mixed multinomial logit model allows for flexible interactions between attributes, the increase in computational complexity is minor, compared to the mixed multinomial logit model. We illustrate the concept and interpretation of the proposed model in a simulation study. We furthermore present a real-world case study from the travel behaviour domain — a bicycle route choice model, based on a large-scale, crowdsourced dataset of GPS trajectories including 119,448 trips made by 8555 cyclists.</div></div>","PeriodicalId":46863,"journal":{"name":"Journal of Choice Modelling","volume":"54 ","pages":"Article 100536"},"PeriodicalIF":2.8,"publicationDate":"2025-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143146713","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

New misspecification tests for multinomial logit models 多项式逻辑模型的新错标检验

IF 2.8 3区经济学 Q1 ECONOMICS

Journal of Choice Modelling

Pub Date : 2025-03-01 Epub Date: 2024-12-12 DOI: 10.1016/j.jocm.2024.100531

Dennis Fok, Richard Paap

Multinomial Logit [MNL] models are misspecified when the Independence of Irrelevant Assumption [IIA] does not hold. In this paper we compare existing tests for IIA with two newly proposed tests. Both new tests use that, when MNL is the true model, preferences across pairs of alternatives can be described by independent binary logit models. The first test compares Composite Likelihood parameter estimates based on pairs of alternatives with standard Maximum Likelihood estimates using a Hausman (1978) test. The second is a test for overidentification in a GMM framework using more pairs than necessary. A detailed Monte Carlo study shows that the GMM test is in general superior with respect to the performance under the null and under the alternative hypothesis. An empirical illustration demonstrates the practical usefulness of the tests.

当不相关假设的独立性[IIA]不成立时，多项Logit [MNL]模型被错误地指定。在本文中，我们比较了现有的IIA测试和两个新提出的测试。当MNL是真实模型时，两个新测试都使用了这样的方法，即通过独立的二元logit模型来描述对备选方案的偏好。第一个测试使用Hausman（1978）测试，将基于对备选方案的复合似然参数估计与标准最大似然估计进行比较。第二个测试是在GMM框架中使用比必要更多的对进行过度标识。一项详细的蒙特卡罗研究表明，GMM检验在零假设和备择假设下的性能一般较好。一个实证说明了这些测试的实际用途。

引用次数: 0

Analysis of attribute importance in multinomial logit models using Shapley values-based methods 基于Shapley值的多项逻辑模型属性重要性分析方法

IF 2.8 3区经济学 Q1 ECONOMICS

Journal of Choice Modelling

Pub Date : 2025-03-01 Epub Date: 2025-01-31 DOI: 10.1016/j.jocm.2025.100538

Patricio Salas , Rodrigo De la Fuente , Sebastian Astroza , Juan Antonio Carrasco

This paper investigates the use of Shapley values-based methods to determine the importance of attributes in discrete choice models, specifically within a Multinomial Logit (MNL) framework. We extend the Shapley decomposition Shorrocks (2013) method from linear models. Additionally, the SHAP method Lundberg and Lee (2017) idea is applied to assess the impact of attributes on individual-level choice probability predictions. A simulation study demonstrates the effectiveness of these approaches under various experimental conditions, including attributes in several ranges and interaction terms. Finally, an empirical application is conducted using well-known travel mode choice datasets. The simulation results show that Shapley values accurately capture the global importance of attributes on goodness-of-fit. The SHAP method provides transparency in MNL model predictions, clarifying how changes in attribute values influence choice probabilities for each decision-maker. These methods offer a complementary perspective to traditional metrics like elasticities and traditional relative importance analysis Orme(2006). In the empirical application, Shapley decomposition highlights the most relevant attributes, while SHAP values uncover individual-level impacts that might not be apparent through elasticities alone. Global and individual-level analysis offers a more comprehensive understanding of attribute importance. In summary, integrating Shapley values with traditional metrics ensures a robust analysis, aiding practitioners and policymakers in making informed decisions based on broad trends and specific impacts.

本文研究了使用基于Shapley值的方法来确定离散选择模型中属性的重要性，特别是在多项式Logit （MNL）框架内。我们从线性模型扩展了Shapley分解Shorrocks（2013）方法。此外，应用SHAP方法Lundberg和Lee（2017）的想法来评估属性对个人层面选择概率预测的影响。仿真研究证明了这些方法在各种实验条件下的有效性，包括几个范围和相互作用项的属性。最后，利用知名的出行方式选择数据集进行了实证应用。仿真结果表明，Shapley值能够准确地捕捉属性在拟合优度上的全局重要性。SHAP方法为MNL模型预测提供了透明度，阐明了属性值的变化如何影响每个决策者的选择概率。这些方法为弹性和传统的相对重要性分析等传统指标提供了补充视角。在实证应用中，Shapley分解突出了最相关的属性，而SHAP值揭示了个人层面的影响，这些影响可能仅通过弹性不明显。整体和个人层面的分析提供了对属性重要性更全面的理解。综上所述，将Shapley价值观与传统指标相结合可确保进行可靠的分析，帮助从业者和政策制定者根据广泛的趋势和具体的影响做出明智的决策。

{"title":"Analysis of attribute importance in multinomial logit models using Shapley values-based methods","authors":"Patricio Salas , Rodrigo De la Fuente , Sebastian Astroza , Juan Antonio Carrasco","doi":"10.1016/j.jocm.2025.100538","DOIUrl":"10.1016/j.jocm.2025.100538","url":null,"abstract":"<div><div>This paper investigates the use of Shapley values-based methods to determine the importance of attributes in discrete choice models, specifically within a Multinomial Logit (MNL) framework. We extend the Shapley decomposition Shorrocks (2013) method from linear models. Additionally, the SHAP method Lundberg and Lee (2017) idea is applied to assess the impact of attributes on individual-level choice probability predictions. A simulation study demonstrates the effectiveness of these approaches under various experimental conditions, including attributes in several ranges and interaction terms. Finally, an empirical application is conducted using well-known travel mode choice datasets. The simulation results show that Shapley values accurately capture the global importance of attributes on goodness-of-fit. The SHAP method provides transparency in MNL model predictions, clarifying how changes in attribute values influence choice probabilities for each decision-maker. These methods offer a complementary perspective to traditional metrics like elasticities and traditional relative importance analysis Orme(2006). In the empirical application, Shapley decomposition highlights the most relevant attributes, while SHAP values uncover individual-level impacts that might not be apparent through elasticities alone. Global and individual-level analysis offers a more comprehensive understanding of attribute importance. In summary, integrating Shapley values with traditional metrics ensures a robust analysis, aiding practitioners and policymakers in making informed decisions based on broad trends and specific impacts.</div></div>","PeriodicalId":46863,"journal":{"name":"Journal of Choice Modelling","volume":"54 ","pages":"Article 100538"},"PeriodicalIF":2.8,"publicationDate":"2025-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143146298","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0