Choice Model Research Articles

Quantitative modelling of cyclists’ route choice behaviour on utilitarian trips based on GPS data: associated factors and behavioural implications

ABSTRACT Despite many years of research and rapid changes in the field of bicycle route choice modelling, a review of factors associated with cyclists’ route choice decisions is missing. This paper reviews 33 studies that employed GPS data and statistical modelling to explain the route choice behaviour of cyclists on utilitarian trips. Factors associated with bicycle route choice are divided into network, contextual, and individual factors. Findings from this review demonstrate that a refined categorisation of model attributes is essential, as the preferences vary greatly between different route attributes. Moreover, the analysis indicates a strong influence of the local context on the attribute definitions and route choice preferences. Although some commonalities between studies w.r.t. the network factors (e.g. a preference towards bicycle infrastructure or avoidance of slopes) can be determined, many results are contradictory or intertwined with other factors. Specifically, results related to intersections, land use characteristics, and interaction with other modes of transport are not always conclusive and are strongly dependent on the local context. However, the explanation of some of the unclear effects can be facilitated by incorporating contextual or individual information in the models. Based on the analysis of the existing studies, this review encourages a joint consideration of studies employing quantitative and qualitative methods to explain cyclists’ behaviour. Moreover, reporting model results in an accessible form (e.g. through relatable values or visualisation) is essential to build bridges between bicycle research and policymaking. Future research is encouraged to study cyclists’ route choice decisions from the perspective of objective well-being, quantifying and modelling the influence of factors such as safety, health, energy expenditure, or noise exposure.

Price Optimization for a Multistage Choice Model

Considering the real-world situations where a customer’s purchase choices in previous stages can influence the prices she encounters in subsequent stages, this research examines the multiproduct price optimization problem under a multistage choice model. Particularly, the seller commits to a multistage pricing policy and determines product prices based on the customer’s purchase history, and the customer makes purchase decisions such that the total expected utility is maximized. We show that the pricing problem has a unique optimal solution under some mild conditions and the optimal solution satisfies a modified equal adjusted markup property. Based on the property, the problem can be solved efficiently by reducing it to a single-dimensional search problem. Moreover, the optimal pricing policy has an important property, namely, the product with a higher adjusted markup in earlier stages should always lead to lower prices in subsequent stages. We also show that compared to customers who are myopic, the seller should offer higher first-stage prices and lower second-stage prices to forward-looking customers, which will lead to a higher profit. Numerical analyses are also conducted to demonstrate the above results.

The active time model of concurrent choice.

The following paper describes a steady-state model of concurrent choice, termed the active time model (ATM). ATM is derived from maximization principles and is characterized by a semi-Markov process. The model proposes that the controlling stimulus in concurrent variable-interval (VI) VI schedules of reinforcement is the time interval since the most recent response, termed here "the active interresponse time" or simply "active time." In the model after a response is generated, it is categorized by a function that relates active times to switch/stay probabilities. In the paper the output of ATM is compared with predictions made by three other models of operant conditioning: melioration, a version of scalar expectancy theory (SET), and momentary maximization. Data sets considered include preferences in multiple-concurrent VI VI schedules, molecular choice patterns, correlations between switching and perseveration, and molar choice proportions. It is shown that ATM can account for all of these data sets, while the other models produce more limited fits. However, rather than argue that ATM is the singular model for concurrent VI VI choice, a consideration of its concept space leads to the conclusion that operant choice is multiply-determined, and that an adaptive viewpoint-one that considers experimental procedures both as selecting mechanisms for animal choice as well as tests of the controlling variables of that choice-is warranted.

What are the determinants of career preferences? The case of high school students in Chile

ABSTRACT Identifying factors influencing career preferences is relevant for faculty, education planners, and authorities when designing educational policies. However, these factors vary according to the socioeconomic context. We estimate a discrete choice model to investigate the determinants of career preferences using a survey of high school students (n = 1,658) in Santiago, Chile. Results show that gender, academic performance, type of school, parents’ occupation, and high school track are significant, but their influences vary among career categories. We concluded that a gender gap exists. Males are more prone to choose STEM areas, whereas females prefer humanities and health careers. The influence of the parents’ occupation is evident in specific fields, with differences between girls and boys. High school vocational track influences career preferences.

PROPERTY RIGHTS GUARANTEES AS PREREQUISITES FOR ECONOMIC COMPETITION: AN EMPIRICAL STUDY

Economic competition by definition implies a choice between competing alternatives. But can economic agents always protect their rights to make this choice? In order to protect property rights, they should have a guarantor who imposes sanctions on the violator. However, the guarantor must be able to use violence, since the imposition of sanctions is often not voluntary. This paper shows that the uneven distribution of violence in society, which is not monopolized by one decision-making center and/or applied outside the rule of law, hinders economic competition. In such a situation, i.e. in a limited access social order, part of property rights remains unprotected, which prevents the entrepreneur from performing his/her function. The reason for this effect is the desire of the elite to maintain the status quo in distribution of rents: they try to avoid changes caused by the market process that may lead to a violation of existing political balance. In this study, the stated theses are confirmed on microdata of the BEEPS industrial enterprises survey by the European Bank for Reconstruction and Development - we test the hypotheses using stochastic frontier analysis (SFA) and discrete choice models. Empirical analysis allows us to argue that in the limited access order, the obstacle to participation in economic competition is a weak property rights protection - it forces firms to compete for limited resources not by satisfying the needs of consumers, but by following the interests of the property rights guarantor. The results of the study can be useful in developing competition policy measures and explaining the phenomenon of «regulatory bias» in antitrust policy of developing countries.

A step in time: A sequence analysis and choice modelling approach to examine time allocation in individual activities and the role of neighbourhood crime

The Space-Time Budget (STB) method is used to collect spatial and temporal features of an individual’s activities and is an important technique to explore relationships between the environment and crime. This spatial and temporal approach to the study of everyday activity spaces reveals much about victimisation experiences, the relationship between place, time, and perceptions of safety, and where and when offending may occur. Few studies, however, consider the relationship between living in a criminogenic place (neighbourhood crime) and people’s daily activity patterns. Drawing on disaggregate app-based data for 50 participants tracked over a 7-day period, we use sequence analysis to first delineate time allocation to each activity on a minute-by-minute basis. Next, using result from the sequence analysis we introduce the multiple discrete-continuous extreme value (MDCEV) model to understand how neighbourhood crime and socio-demographic characteristics influence individuals’ time allocation to discrete daily activities. Results reveal that neighbourhood drug and violent incidents exert restrictions on an individual’s propensity to spend time outside the home. Females and older people appear more likely to be constrained by the presence of neighbourhood crime (in particular drug and property incidents) as they allocate less time to outdoor activities. The principal utility of the current study is its methodological advancement and the practical insights for urban planning regarding the design of crime prevention strategies to increase guardianship in public places.

Identification and Estimation of Discrete Choice Models with Unobserved Choice Sets

We propose a framework for nonparametric identification and estimation of discrete choice models with unobserved choice sets. We recover the joint distribution of choice sets and preferences from a cross-section of repeated choices. We assume that either the latent choice sets are sparse or that the number of repeated choices is sufficiently large. Sparsity requires the number of possible choice sets to be relatively small. It is satisfied, for instance, when the choice sets are nested or when they form a partition. Our estimation procedure is computationally fast and uses mixed-integer programming to recover the sparse support of choice sets. Analyzing the ready-to-eat cereal industry using a household scanner dataset, we find that ignoring the unobservability of choice sets can lead to incorrect estimates of preferences.

REAL ANALYTIC DISCRETE CHOICE MODELS OF DEMAND: THEORY AND IMPLICATIONS

We demonstrate that a large class of discrete choice models of demand can be approximated by real analytic demand models. We obtain this result by combining (i) a novel real analytic property of the mixed logit and the mixed probit models with any distribution of random coefficients and (ii) an approximation property of finite mixtures of Gumbel and Gaussian distributions. To illustrate some of the implications of this result, we discuss how real analyticity facilitates nonparametric and semi-nonparametric identification, extrapolation to hypothetical counterfactuals, numerical implementation of demand inverses, and numerical implementation of the maximum likelihood estimator.

Probing the limit of hydrologic predictability with the Transformer network

For a number of years since their introduction to hydrology, recurrent neural networks like long short-term memory (LSTM) networks have proven remarkably difficult to surpass in terms of daily hydrograph metrics on community-shared benchmarks. Outside of hydrology, Transformers have now become the model of choice for sequential prediction tasks, making it a curious architecture to investigate for application to hydrology. Here, we first show that a vanilla (basic) Transformer architecture is not competitive against LSTM on the widely benchmarked CAMELS streamflow dataset, and lagged especially prominently for the high-flow metrics, perhaps due to the lack of memory mechanisms. However, a recurrence-free variant of the Transformer model obtained mixed comparisons with LSTM, producing very slightly higher Kling-Gupta efficiency coefficients (KGE), along with other metrics. The lack of advantages for the vanilla Transformer network is linked to the nature of hydrologic processes. Additionally, similar to LSTM, the Transformer can also merge multiple meteorological forcing datasets to improve model performance. Therefore, the modified Transformer represents a rare competitive architecture to LSTM in rigorous benchmarks. Valuable lessons were learned: (1) the basic Transformer architecture is not suitable for hydrologic modeling; (2) the recurrence-free modification is beneficial, so future work should continue to test such modifications; and (3) the performance of state-of-the-art models may be close to the prediction limits of the dataset. As a non-recurrent model, the Transformer may bear scale advantages for learning from bigger datasets and storing knowledge. This work lays the groundwork for future explorations into pretraining models, serving as a foundational benchmark that underscores the potential benefits in hydrology.

Dually flat structure of binary choice models

AbstractIn this study, we consider parametric binary choice models from the perspective of information geometry. The set of models is a dually flat manifold with dual connections, naturally derived from the Fisher information metric. Under the dual connections, the canonical divergence and the Kullback–Leibler divergence of the binary choice model coincide if and only if the model is a logit model.

Online Learning for Constrained Assortment Optimization Under Markov Chain Choice Model

Assortment optimization finds many important applications in both brick-and-mortar and online retailing. Decision makers select a subset of products to offer to customers from a universe of substitutable products, based on the assumption that customers purchase according to a Markov chain choice model, which is a very general choice model encompassing many popular models. The existing literature predominantly assumes that the customer arrival process and the Markov chain choice model parameters are given as input to the stochastic optimization model. However, in practice, decision makers may not have this information and must learn them while maximizing the total expected revenue on the fly. In “Online Learning for Constrained Assortment Optimization under the Markov Chain Choice Model,” S. Li, Q. Luo, Z. Huang, and C. Shi developed a series of online learning algorithms for Markov chain choice-based assortment optimization problems with efficiency, as well as provable performance guarantees.

Assessing the market opportunities for pesticide-free vegetables in Bangladesh: consumers’ preferences and willingness to pay

The excessive use of fertilizers and pesticides in agriculture poses threats to ecosystems and human health. To mitigate these risks, researchers have turned their attention to pesticide-free production methods. This study aimed to assess the market opportunities for pesticide-free vegetables (PFVs) through determining consumers’ preferences and willingness to pay (WTP), along with identifying factors influencing consumers' WTP decision in three major cities of the country, namely Dhaka, Khulna, and Mymensingh. Employing a multistage cluster sampling method, 150 samples were selected and data collection utilized the android-based app Kobo through direct interviews. The contingent valuation method (CVM) and double-bounded dichotomous choice model (DBDC) were used to estimate WTP and preferences while ordered logit model were employed to identify influencing factors. The findings indicated that over 90% of consumers in Mymensingh and over 92% in Dhaka and Khulna were willing to pay a premium for pesticide-free tomatoes and brinjal. Additionally, more than 18% and 19% of consumers expressed willingness to pay a 20% premium for brinjal and tomato, while over 39% were willing to pay for both a 10% premium. The study revealed that factors such as age, sex, education, household size, and income significantly influenced consumers’ WTP decisions, with age showing a negative correlation, while education and income were positively associated with interest in consuming pesticide-free vegetables. Overall, the study underscores the increasing demand for pesticide-free vegetables in urban areas, emphasizing the need for governments and stakeholders to develop supportive policies for sustainable market development. Res. Agric. Livest. Fish. Vol. 11, No. 1, April 2024: 69-77

Do Health and Housing Attributes Motivate Residential Moves Among Older Chinese Adults? Evidence From an 8‑Year Follow‑up Study.

Despite the widespread belief in aging-in-place as the preferred housing arrangement for older adults, they are increasingly embracing moving. The person-environment fit theory explains that environmental changes due to residential moves can pose health risks, discouraging older people from moving. However, it also suggests that moves may be suitable when living conditions no longer meet their physical needs. The correlation between older people's health, housing, and their subsequent moving or staying actions in China remains underexplored. Using alternative operating variables of key concepts and the China Family Panel Studies 2012-2018 data sets, this study examines the effects of health and housing status on older people's residential moves in China. The study outlines changes in health indicators and housing characteristics during multiple moves, as well as examines the relationship between the health and housing status of older adults and moving over a relatively long period of time using both regression models with lagged explanatory variables and fixed effects binary choice models. The results reveal that: (1) higher rates of subsequent moves were observed among older adults with better self-rated health, positive self-rated health changes, and no ADL impairment, but no significant associations were found between most health variables and moving; and (2) the correlation between older persons' house ownership/type and their residential moves was significant and consistent over time, steady and lasting. Potential mechanisms explaining the association between specific housing types and ownership statuses on moving are discussed. The findings encourage a focus on the positive benefits of moving in later life and how to provide additional housing options for older individuals.

Mesoscopic model of cycling trip energy expenditure based on operating modes

Cyclist physical exertion is largely ignored in quantitative travel analysis, partly due to a lack of appropriate tools. Microscopic models of second-by-second energy expenditure based on equations of motion are data intensive and cannot be applied to hypothetical routes (such as needed for route choice modelling). Macroscopic models of aggregate energy expenditure based on a fixed assumed energy intensity are insensitive to traveller, trip, and contextual factors that are relevant for behavioural research and policy analysis (such as bicycle type or trip purpose). Building on concepts from motor vehicle emissions analysis, this paper proposes a mesoscopic approach to model cycling trip energy expenditure based on the distribution of travel time across discrete states of motion (“operating modes”) for different classes of traveller and trip (“model segments”). We aim to answer two key questions for model implementation: 1) which variables most effectively classify trips into model segments and 2) what operating mode definition most consistently characterizes cycling energy expenditure within model segments? We also evaluate the precision of the mesoscopic model relative to cycling energy estimates from microscopic and macroscopic models. Applied to a dataset of naturalistic cycling trips in Vancouver, Canada, the proposed mesoscopic model with six model segments based on 3 segmenting variables (rider gender, electric-assist bicycle, and high or low speed tier) explains up to 28 % of the variance in trip-level energy estimates from the microscopic model (within around 35 W of the microscopic estimates, on average). Further research to develop cycling trip energy models for general application is discussed.

Differences in Vaccination Consultation Preferred by Primary Health Care Workers and Residents in Community Settings.

To evaluate the preference of primary HCWs and residents on vaccination consultation in community health services to provide evidence for vaccine hesitancy intervention strategies. A discrete choice model (DCM) was constructed to evaluate the preference difference between primary HCWs and residents on vaccination consultation in community health services in China during May-July 2022. A total of 282 residents and 204 HCWs were enrolled in this study. The residents preferred consulting with an HCW-led approach (β = 2.168), with specialized content (β = 0.954), and accompanied by telephone follow-up (β = 1.552). In contrast, the HCWs preferred face-to-face consultation (β = 0.540) with an HCW-led approach (β = 0.458) and specialized content (β = 0.409), accompanied by telephone follow-up (β = 0.831). College residents and residents with underlying self-reported disease may be near-critically inclined to choose traditional consultation (an offline, face-to-face consultation with standardized content and more prolonged duration) rather than a new-media consulting group (an online consultation with specialized content within 5 min). Urban HCWs preferred long-term consultation groups (the resident-led offline consultation with follow-up lasting more than 5 min). In contrast, rural HCWs preferred efficient consultation (the HCW-led, short-duration, standardized offline consultation mode). The selection preference for vaccine consultation reveals a gap between providers and demanders, with different groups exhibiting distinct preferences. Identifying these targeted gaps can help design more acceptable and efficient interventions, increasing their likelihood of success and leading to better resource allocation for policymakers to develop targeted vaccination policies.

I Would Like to but I Cannot: What Influences the Involuntariness of Part-Time Employment in Italy

Over the last two decades, involuntary part-time (IPT) employment has become a more and more pressing issue in Europe, especially in the southern countries, where IPT today constitutes most part-time employment. Using INAPP-PLUS data and different discrete choice model estimations, this paper aims to shed light on the factors that explain the IPT growth in Italy, focusing on what influences the IPT status at the individual, household and labour market levels. The main hypothesis is that what influences the IPT work derive from a combination of workers’ individual, household, and job characteristics which may engender limited power during the bargaining process. The empirical results, based on gender-specific models, highlight that characteristics associated with the IPT status significantly changed over time, reporting a convergent path between the gender profiles of IPT employment. However, IPT employment for women still appears to be mainly originated from the gendered division of domestic and care tasks, while this phenomenon seems to be mainly driven by the labour demand side for men.

Development and Evaluation of Mycoplasma gallisepticum Challenge Model in Layer Pullets.

Manufacturers of Mycoplasma gallisepticum (MG) modified live vaccines usually recommend a single application at 8 wk of age. This makes 12-16-wk-old layer pullets suitable for challenge studies intended to evaluate these vaccines. Numerous challenge models in different poultry species and ages have been reported. However, there is not an established layer pullet challenge model for this age. The aim of this study is to develop a suitable challenge model in 12-wk-old layer pullets. MG Rlow strain was used as the challenge strain, and its ability to induce clinical signs and lesions in 12-wk-old Hy-Line W-36 layer pullets was evaluated. Three different doses (low, 7.95 × 104 color-changing units [CCU]/bird; medium, 7.95 × 106 CCU/bird; and high, 7.95 × 108 CCU/bird) via three different routes (eye drop, fine spray, and contact infection) were compared and evaluated using different parameters. At 14 days post-challenge, there were no mortalities in any of the groups throughout the study. Layer pullets directly challenged with the high dose via the fine spray route showed the clearest and most consistent results (clinical signs, positive quantitative real-time PCR [qPCR], seroconversion, air sac scoring, and histopathological changes of the tracheal mucosa). Medium and low challenge doses applied via fine spray or eye drop did not show consistent results. Rlow strain was able to spread to the contact infection birds, as confirmed by the positive qPCR results; however, none of the contact-infected birds showed any clinical signs or gross or microscopic lesions. Our results suggest that a high dose (7.95 × 108 CCU/bird) administered through a fine spray route is the model of choice in any future MG vaccine evaluation trials in 12-wk-old layer pullets.

Impact of the Local Dynamics on Exit Choice Behaviour in Evacuation Model

This study investigated the interplay between exit selection models and local pedestrian movement patterns within floor field frameworks. Specifically, this investigation analysed the performance of a multinomial logit exit choice model, incorporating both expected utility theory and cumulative prospect theory frameworks when coupled with three distinct local-level pedestrian movement models (FF-Von Neumann, FF-Moore, and NSFF). The expected utility theory framework considers the deterministic component as a linear relationship, while the cumulative prospect theory framework further considers the decision-maker’s risky attitudes by transforming objective terms into subjective terms using a power value function. The core objective was to comprehend how local movement dynamics, as represented by the floor field models, influence decision-making during exit selection. Comparative analyses revealed intriguing variations between the three local models, despite their shared expected utility theory-based exit choice framework. These discrepancies stemmed from the diverse pedestrian trajectory behaviours generated by each model. Consequently, these local dynamics impacted the decision-maker’s assessment of critical factors, such as the number of evacuees close to the decision-maker (NCDM) and the number of evacuees close to an exit (NCE), which the exit choice model incorporates. These assessments, in turn, significantly affected higher-level decision-making. The integration of the three models with the multinomial logit exit choice model, using either cumulative prospect theory and expected utility theory frameworks, further strengthened the observed bilateral relationship. While the specific nature of this relationship varied depending on the chosen framework and its implementation details, these consistent findings demonstrate the robustness of the results. This reinforced the influence of local-level pedestrian dynamics on higher-level exit selection, highlighting the importance of accurate crowd dynamics modelling, especially when advanced exit choice models consider local movement factors.

Reinforcement learning-based order-dispatching optimization in the ride-sourcing service

Since the emergence of ride-sourcing, the dispatching problem has been a hot research issue in the travel market. Previous studies designed various refined algorithms to improve the dispatching efficiency. However, the heterogeneity of passenger cancellation behavior and driver's work patterns is ignored in the modeling of these studies, which may result in the algorithms failing to entirely capture the randomness in the real world, thereby affecting dispatching efficiency. To fill this research gap, this paper employs a discrete choice model and Gaussian mixture model to model passenger cancellation behavior and extract driver working patterns, respectively. Furthermore, a cooperative deep deterministic policy gradient (DDPG) algorithm is proposed for dispatching. The algorithm constructs decision networks separately for drivers with different working patterns to learn the optimal dispatching decision and designs a centralized neural network to coordinate the dispatching of all vehicles to maximize the global benefit. A price incentive strategy is proposed to improve passenger loyalty by providing incentive prices. Moreover, based on the constructed discrete choice model, a simulation environment is designed to describe the cancellation behavior in the real world and support the training of the algorithm. Experimental results based on real data in Beijing show that the cooperative DDPG algorithm can improve the order response rate by 2% to 6% and platform revenue by 3% to 9% in multiple travel periods while providing a better travel experience to passengers. Additionally, the price incentive strategy also effectively reduces passenger cancellation.

A Comparative Study between Time Series and Machine Learning Technique to Predict Dengue Fever in Dhaka City

The dengue virus is the most dangerous one that mosquitoes may spread to people. Despite attempts by the government, dengue outbreaks are becoming increasingly common in Bangladesh. Interventions in public health rely heavily on KAP (knowledge, attitude, and practice) studies. The primary goal of this research is to forecast the occurrence of dengue disease in the city of Dhaka using methods from machine learning and time series analysis and then to compare the models in order to find the one with the lowest MAPE. From January 2016 through July 2021, monthly data were retrieved for this study from WHO and the Directorate General of Health and Services (DGHS). According to the findings of this research, neural networks outperform time series analysis when it comes to making predictions. The best-fitted neural network (NN) model was found in model 04 with 05 hidden layers which produced the minimum error model with the value of error 0.003032557, and the values of RMSE and MAPE are 7.588889e − 06 and 1.15273, respectively, for the prediction of the dengue fever in Dhaka city. In contrast, the original dengue data in the time series analysis is not stationary. Take the difference and run the unit root test by the augmented Dickey–Fuller (ADF) test to make it stationary. The dengue data series is stationary at the first-order difference, as evidenced by the ACF and PACF, which show no noticeable spike in the first-order difference. The ARIMA (6, 1, 1) model with the lowest AIC = −251.8, RMSE = 0.0310797, and MAPE = 15.2892 is the best choice model for predicting the dengue death rate. Therefore, from these two models, the NN model gives better prediction performance with the lowest value of MAPE. So, the neural network gives better prediction performance than time series analysis. The NN model forecasted 12-month death rates of dengue fever that suggest the death rate in dengue fever falling month by month. This study is more innovative than any other research because this research approach is different from any other research approach. The model selection criteria are based on the most effective performance metrics MAPE, indicating the lowest error and better prediction performance. Therefore, from this research, the author suggests machine learning gives better prediction performance than time series analysis for any other prediction performance.