Multinomial Probit Research Articles

Mixed Multinomial Probit Model Accommodating Flexible Covariance Structure and Random Taste Variation: An Application to Commute Mode Choice Behavior

This paper developed a mixed multinomial probit (MMNP) model with alternative error specification and random coefficients (for both generic variables and personal attributes) to accommodate flexible covariance structure and taste variation. The MMNP model can be efficiently estimated with analytic approximations of multivariate normal cumulative distribution functions, which avoid defects of simulation-based integration in the mixed multinomial logit (MMNL) model. The integral dimension of the MMNL model increases as random coefficients increase, but it only depends on the number of available alternatives in the MMNP model. Simulation experiments and empirical analysis of Shanghai commuters’ mode choice behavior were undertaken to examine the performance of MMNP models. Both simulation results and empirical results show that MMNP models can well accommodate flexible covariance structures and taste variation reflected through random coefficients being associated with both generic and personal variables. Empirical results indicate that the MMNP model performs better than traditional discrete choice models, such as the multinomial logit, the cross-nested logit, MMNL, and multinomial probit models. Random coefficients of “in-vehicle time of car” and “number of companions” indicate taste heterogeneity and the identifiability of random coefficients associated with both generic and personal attributes. Pairwise positive correlations between car/taxi, bus/metro, and bus/bus and metro are to be expected. However, the positive correlation between the car and metro modes may be unique to the Chinese city, Shanghai, because of the developed metro system. Unequal error variances reflect heterogeneities in unspecified factors in commute modes’ utilities. The MMNP model will offer an alternative efficient way to accommodate taste heterogeneity and flexible error covariance structure in discrete choice models. Compared with the MMNL model, the MMNP model can accommodate more random coefficients without increasing computational complexity.

Impact of Family Background on Cognitive Skills: An Empirical Investigation

This study desires to look at the impact of family background on cognitive skills of the children. On the basis of primary survey of two districts of West Bengal—namely Bankura and Nadia—and using the fathers’ education, occupation and per capita household income as the proxies for family background we have employed multinomial probit and ordered logit regression model for this analysis. The findings reveal that there exists a significant positive association between family background and cognitive skills of the child indicating the existence of poor intergenerational mobility in education in the surveyed region. JEL Codes: I20, I24, I28

ADAPTATION TO CLIMATE CHANGE IN ARID LANDS: EVIDENCE FROM PASTORAL AREAS OF SENEGAL

This paper analyzes the determinants of adaptation options in pastoral drylands and investigates whether adaptation strategies can be used jointly. We assume that decisions can be made jointly as complements or substitutes and investigates whether herders in Senegal adapt to climate change by pursuing multiple strategies. We use a multinomial probit model with primary data collected from 410 herders of Senegalese drylands to identify adaptation determinants. Results show that 73.7% of the surveyed households rely on at least one adaptation strategy including storage of livestock feed, increased mobility, changes in water management, diversification of activities and changes in herd composition. Moreover, we notice that adaptation decisions of pastoral households can be taken jointly and those with mobility do not pursue other adaptation strategies, while those lacking mobility undertake multiple strategies. The diversity of factors explaining adaptation calls for targeted policies that promote adaptation strategies to strengthen the resilience of pastoralists.

Flood disaster risk perception and evacuation willingness of urban households: The case of Accra, Ghana

Flooding is Ghana's most occurrent natural disaster, with Accra, the capital city and central economic hub experiencing the highest flood incidence and the effects on lives and properties. With the increasing trend of data on rainfall above 20 mm (heavy rain) in Accra from 1980 to 2020, a high possibility of future flood hazards has been projected, with the frequency and intensity of flood occurrences expected to increase. Thus, this study aimed to analyze the flood disaster risk perception of urban households in Accra and how it influences their willingness to leave the flood-prone zone. The multinomial probit model was used to analyze the evacuation willingness of households, and binary logit and probit models were also used to test the robustness of the estimates. The results indicate that the odds of being flood risk lovers are found to be decreased by the number of years of formal education, gender, household level of income, physical mobility, costs incurred from previous floods, probability and severity of future floods, and perception about taxing building in flood zones. The odds of being flood risk lovers are increased by household duration in the flood zone, size of household, the experience of past flood occurrences, close distance to work, and residence close to Odaw river/Korle lagoon. The study suggests that policies should target increasing public awareness of flood risks, providing affordable housing for residents, banning new settlements on flood-prone zones, and building infrastructure to mitigate flood risks.

Financial inclusion and households’ choice of solid waste disposal in Ghana

BackgroundAs the amount of solid waste generated by households in Ghana continues to grow, policy makers are preoccupied with finding better means of managing these solid wastes. To this end, a number of studies have been conducted on the factors that determine the choice of solid waste disposal method among households in Ghana. Notwithstanding, while financial inclusion is deemed as an effective tool for improved solid waste management, none of these studies paid attention to it. This study therefore, investigates the effect of financial inclusion on the choice of solid waste disposal method among households in Ghana.MethodsThe study uses data from the Ghana Living Standards Survey round 7 (GLSS7). The multinomial probit regression is used as the empirical estimation technique.ResultsOur results show that financial inclusion increases the likelihood of households opting for the collection method of solid waste disposal relative to burning, public dumping and indiscriminate disposal of solid waste.ConclusionFinancial inclusion enables households to opt for a healthy solid waste disposal method (collection method), hence, in policy makers’ attempts to improve solid waste disposal, paying attention to financial inclusion can be a very useful strategy.

Which hospital workers do (not) want the jab? Behavioral correlates of COVID-19 vaccine willingness among employees of Swiss hospitals.

In many countries, the current vaccination rates are stagnating, to the extent that vaccine hesitancy—the delay or refusal to take recommended vaccinations—forms a major obstacle to ending the COVID-19 pandemic. This tendency is particularly concerning when observed among healthcare workers who are opinion leaders on medical matters for their patients and peers. Our study surveys 965 employees of two large Swiss hospitals and profiles vaccine-hesitant hospital employees using not only socio-demographic characteristics, but also a comprehensive set of standard behavioral preference measures: (i) Big-5 personality traits, (ii) risk-, time- and social preferences, and (iii) perceived prevailing social norms. Using multinomial probit models and linear probability models, we find that vaccine-hesitant hospital employees are less patient and less likely to perceive vaccination as the prevailing social norm—in addition to replicating previously published socio-demographic results. Our findings are robust to a range of model specifications, as well as individual and situational covariates. Our study thus offers actionable policy implications for tailoring public-health communications to vaccine-hesitant hospital employees.

Understanding participation in non-farm activities in rural Sudan: across-sector study of irrigated and rainfed agricultural systems

PurposeThis paper examines the determinants of participation in non-farm activities in rural Sudan. It also investigates whether the factors that influence participation in non-farm activities vary across agriculture sub-sectors.Design/methodology/approachThe study adopts the multinomial logit and probit methods on labour supply participation theory using the Sudanese National Baseline Household Survey (2009). The analysis was applied across job types and agriculture sub-sectors.FindingsThe results indicated that educational level, means of transportation, lack of land and access to formal agricultural credit are the most significant factors that push rural farmers to participate in non-farm activities. Surprisingly, the effect of household income was positive and significant, implying that individuals from rich households have higher opportunity to engage in non-farm activities compared to their poor counterparts. The results also revealed some variations in the factors that influence participation in non-farm activities according to the agricultural sub-sectors.Originality/valueThe originality of this article lies in investigating the factors that influence participation in non-farm activities across irrigated and rainfed systems. To the best of our knowledge, this is the first study identifying the determinants of participation in non-farm activities across agriculture sub-sectors. Therefore, the paper fills an important gap in the literature and helps in designing appropriate pro-poor policies to allocate infrastructures across irrigated and rainfed areas in Sudan.Peer reviewThe peer review history for this article is available at https://publons.com/publon/10.1108/IJSE-02-2022-0092.

Modeling the Impact of Weather and Context Data on Transport Mode Choices: A Case Study of GPS Trajectories from Beijing

Over the years, researchers have been studying the effects of weather and context data on transport mode choices. Existing research studies are predominantly designed around travel surveys, but the accuracy of their findings relies on how travelers give accurate and honest answers. The proliferation of smartphones, however, now offers the possibility of utilizing GPS positioning data as an alternative information source, opening the potential to accurately model and better understand factors which influence transport mode choices, compared to travel surveys. The objective of this work is to develop a model to predict the transport mode choices based on GPS trajectories, weather and context data. We use 2671 GPS trajectories from the Geolife GPS trajectories dataset, weather data, such as temperature and air quality, and context data, such as rush hour, day/night time and onetime events, such as the Olympics. In the statistical analysis, we apply both descriptive and statistical models, such as the multinomial logit and probit models. We find that temperature has the most prominent effect among weather conditions. For instance, for temperatures greater than 25 °C, the walking share increases by 27%, and the bike share reduces by 21%, which is line with the results from several survey-based studies. In addition, the evidence of government policy on transport regulation is revealed when the air quality becomes hazardous, as people are encouraged to use environmentally friendly transport mode choices, such as the bike instead of the bus or car, which are known CO2 emitters. Our conclusion is that GPS trajectories can be used as a means to model passenger behavior, e.g. the choice of transport mode, in a quantitative way, which will support transport mode operators and policy makers in their efforts to design and plan the transport mode infrastructure to best suit the passengers’ needs.

Synchronized agricultural credit and diversification adoption to catastrophic risk manage for wheat production in Punjab, Pakistan.

In developing countries, particularly in agrarian economies like Pakistan, agricultural production is severely affected by successive incidences of adverse catastrophic risks. Inadequate and limited risk management measures from farming community to overcome such severe financial, production, and marketing risks not enough so need of adoption farm-based feasible risk management strategies. In addressing this research gap, this study investigated the potential association and implementing synchronized agricultural credit and diversification adoption to catastrophic risk manage for wheat production in Punjab, Pakistan. The study used the data of 480 wheat farmers' respondents and employed multinomial and bivariate probit regression models for empirical analysis. Empirical estimates illustrated the significant association in adoption of multiple risk management strategies as addressing the adoption single risk management strategy motivates farmers to adopt another strategy at same time. Furthermore, results also indicated as farmer schooling, age, family return, ownership of land, farm size, and farmer risk oppose character were highlighted the most influenced features for adopting various risk management strategies. Policy makers and state-based authorities can be assisted by these outcomes to evaluating plans of risk management and willingness of farmer in accept government supported regarding implementation of these catastrophic risk managing strategies.

Local perception and attitude toward community forest concessions in the Democratic Republic of Congo

The effects of community forestry on biodiversity conservation and local development have been highlighted in Africa and elsewhere. However, little is known on the perception and attitude of local forest users toward community forestry programs. This study aimed to address this gap by identifying the factors that influence the perception and attitude of local forest users toward Local Community Forest Concessions (LCFCs) in the Democratic Republic of Congo (DRC). A survey of 138 households was conducted in three villages of the Bisemulu LCFC in eastern DRC. The responses to the close-ended questions were analyzed with a multinomial Probit regression to determine the factors explaining LCFC perception. The results revealed that the key factors determining positive attitude toward the LCFC include education, age, occupation, knowledge of the legal framework and family size. However, challenges remain to the local acceptability of the LCFC, as government employees, elders, youth and larger households expressed negative attitudes. Awareness campaigns mobilizing civil society organizations that have higher level of trust in the communities may overcome some of the challenges. Administrative procedures and the associated costs were seen as major obstacles to LCFC implementation and operation. Procedure simplification and funding support to the LCFCs would thus be needed. The results of this study will contribute to the formulation of sustainable forestry policies to make LCFCs a lever for local development in the DRC and elsewhere.

Efficient Planning and Selection of Media Advertising Using Linier Programming and Machine Learning Methods

Due to rapid development of information technology and drastic innovations in technology that change customers' media use behavior, it is necessary to analyze the data and make systematic predictions in media planning and selection. Our research used linear programming method to plan and select media and put forward the most effective way to deliver advertising to the most targeted users at the lowest cost. In this study we used data from 2,451 households in 9 districts of Ulaanbaatar, 17 local aimags*, and 26 soums* in the 2020 National Syndicate Survey to determine the lifestyle of Mongolians, as well as data from more than 300 media outlets of 9 types, including such as television, internet, and social media. The study examined the factors that influence amount of time spent on television, website, social media, and FM radio using multinomial logit and multinomial probit models. However, the fact that consumers spend a lot of time on a media outlet does not mean that their purchasing decision will be affected. Therefore, we used the feature selection method based on 16 indicators to determine which media channel ad the customer viewed and made a purchase. We analyzed media users based on 54 data mining and classification indicators, including age, gender, and location. As a result of the study we developed a framework, highlighted the national media use behavior, and proposed a media planning methodology for businesses. The principal author of the article Mr. Amgalanbaatar Dagdandorj is a doctoral student at the Department of Marketing and International Trade of the Business School, National University of Mongolia. He is a CEO of the research and consultation company Mongolian Marketing Consulting group. He developed a research methodology for the National media use syndicate survey and received the copyright certificate. The current research analyzed the survey data implemented during the last eight consecutive years in Mongolia within the dissertation framework of “Efficient planning and selection of media advertising using linear programming and machine learning methods” as a requirement for a degree of Doctor in Business Administration. As for this work, D. Amgalanbaatar, as the principal author of the article, worked on the research data analysis, developed a methodology and modeling as well as media planning model and recommendations for its application. The last author of this article, Dr. Batnasan Namsrai, works as a Professor at the Marketing and International Trade Department of the Business School, National University of Mongolia. He is the academic advisor of the doctoral student and worked as a consultant in developing the methodology and design of the National media use syndicate survey. As the last author of the current article, he formulated a general methodology and developed the logic and format of a scientific article. The findings and the planning model developed on its basis can be the primary source for effective planning of the country's marketing and advertising activities of Mongolian business, government and media organizations.

The innovation and exports interplay across Africa: Does business environment matter?

The paper investigates the relationship between innovation and export behaviour across manufacturing and services firms in Africa. The study is based on the general premise that innovation has a positive effect on firm’s exports (self-selection hypothesis) and the complementary assumption that internationalisation drives firms to innovate (learning-by-exporting hypothesis). To test this complex, two-way link between innovation and exports, the study contributes to the existing literature by analysing the complementarity effects between product and process innovation in their relationship with exports. A combination of process and product innovation is expected to have a greater impact on the likelihood of firms entering the foreign market and on their export performance. Using data from 45 African countries, from 2006 to 2020, the multinomial probit and two-stage least squares models are estimated. There is support for the learning by exporting and the self-selection hypotheses for African firms. The findings also reveal the need to improve the business environment across African economies to foster greater exports and innovation.

A multinomial probit model with Choquet integral and attribute cut-offs

Several non-linear functions and machine learning methods have been developed for flexible specification of the systematic utility in discrete choice models. However, they lack interpretability, do not ensure monotonicity conditions, and restrict substitution patterns. We address the first two challenges by modeling the systematic utility using the Choquet Integral (CI) function and the last one by embedding CI into the multinomial probit (MNP) choice probability kernel. We also extend the MNP-CI model to account for attribute cut-offs that enable a modeler to approximately mimic the semi-compensatory behavior using the traditional choice experiment data. The MNP-CI model is estimated using a constrained maximum likelihood approach, and its statistical properties are validated through a comprehensive Monte Carlo study. The CI-based choice model is empirically advantageous as it captures interaction effects while maintaining monotonicity. It also provides information on the complementarity between pairs of attributes coupled with their importance ranking as a by-product of the estimation. These insights could potentially assist policymakers in making policies to improve the preference level for an alternative. These advantages of the MNP-CI model with attribute cut-offs are illustrated in an empirical application to understand New Yorkers’ preferences towards mobility-on-demand services.

Factors Influencing the Choice of Storage Technologies by Smallholder Potato Farmers in Eastern and Southwestern Uganda

Potato is a key food and cash crop in Uganda, mainly produced by smallholder farmers in the eastern and southwestern highlands of the country. This study assessed different factors influencing the choice of storage technologies by Ugandan potato farmers. Data were collected from 240 potato farmers using structured questionnaires in key potato producing districts in eastern and southwestern Uganda. Data were analysed using descriptive statistics and the multinomial probit regression model. Results indicate that potato farmers have limited access to credit and adequate extension services. Furthermore, most of the potato production is sold immediately after harvest. Although significant quantities of potato are stored as food for the household and seed for the next season, very few farmers store ware potato for later sale at a higher price. The farmer households generally use light storage technologies designed for seed storage, while dark stores required for proper ware potato storage are rarely used. Results for factors influencing the choice of storage technologies were mixed, and the extent and direction of influence varied with technology. The predominant factors that positively influenced the choice of dark storage technologies or a combination of different storage technologies included monthly income from sources other than potato sales, access to storage management advice and access to credit. This study recommends enhancing farmers’ access to adequate extension services and credit to promote good ware potato storage conditions.

Households’ fuel choices in the context of a decade-long load-shedding problem in Nepal

Access to an uninterrupted electricity supply is an important driver of growth and human development. However, people in many developing countries do not yet have access to a steady electricity supply – partly due to power cuts, blackouts, and load-shedding. Such disasters exert pressure on households and hinder energy modernization. This paper investigates how households' fuel choices are possibly distorted amidst unreliable electricity supply. It studied households' cooking fuel choices and fuel switching behavior in the face of the load-shedding problem in Nepal (2006–2016). This study employs two National Living Standard Surveys (NLSSs) and four Annual Household Surveys (AHSs). Multinomial Logit and Multinomial Probit model estimations suggest that the crisis delayed households' energy modernization drive. Furthermore, households diversified their fuel basket and increased ‘energy stacking’ in response to the crisis. Despite a trend of increasing preferences towards modern fuels, significant reliance on traditional fuels remains. Therefore, it is recommended to improve supply reliability remarkably for a successful transition toward modern energy. Moreover, the nature of the crisis needs to be well understood to formulate efficient policies on energy transition since different crises might affect households differently.

The psychological outcomes of COVID-19 affected the pandemic-after risk perceptions of nurse clinicians: a latent profile analysis.

Risk perception among nurses after the COVID-19 pandemic is a crucial factor affecting their attitudes and willingness to work in clinics. Those with poor psychological status could perceive risks sensitively as fears or threats that are discouraging. This article aimed to determine whether psychological outcomes, including post-traumatic stress disorder (PTSD), depression, anxiety, and insomnia, following the COVID-19 pandemic were differentially related to the risk perceptions of nurses working in clinics and increased perceived risk. The participants were 668 nurse clinicians from five local hospitals. Risk perceptions and psychological outcomes were measured by adapted questionnaires via the Internet. Latent profile analysis (LPA) identified subgroups of individuals who showed similar profiles regarding the perceived risks in nursing. Multinomial regression and probit regression were used to examine the extent to which sociodemographic and psychological outcomes predicted class membership. LPA revealed four classes: groups with low-, mild-, moderate-, and high-level risk perceptions. Membership of the high-level risk perception class was predicted by the severity of psychological outcomes. Anxiety significantly accounted for a moderate increase in risk perceptions, while the symptoms of insomnia, depression, and PTSD accelerated the increase to the high level of risk perception class. By classifying groups of nurse clinicians sharing similar profiles regarding risk perceptions and then exploring associated predictors, this study shows the psychological outcomes after COVID-19 significantly impacted pandemic-associated risk perceptions and suggests intervening in nurses' psychological outcomes while simultaneously focusing on work-related worries is important following the outbreak of COVID-19.

SPARSE TENSOR PRODUCT APPROXIMATION FOR A CLASS OF GENERALIZED METHOD OF MOMENTS ESTIMATORS

Generalized Method of Moments (GMM) estimators in their various forms, including the popular Maximum Likelihood (ML) estimator, are frequently applied for the evaluation of complex econometric models with not analytically computable moment or likelihood functions. As the objective functions of GMM- and ML-estimators themselves constitute the approximation of an integral, more precisely of the expected value over the real world data space, the question arises whether the approximation of the moment function and the simulation of the entire objective function can be combined. Motivated by the popular Probit and Mixed Logit models, we consider double integrals with a linking function which stems from the considered estimator, e.g. the logarithm for Maximum Likelihood, and apply a sparse tensor product quadrature to reduce the computational effort for the approximation of the combined integral. Given Holder continuity of the linking function, we prove that this approach can improve the order of the convergence rate of the classical GMM- and ML-estimator by a factor of two, even for integrands of low regularity or high dimensionality. This result is illustrated by numerical simulations of Mixed Logit and Multinomial Probit integrals which are estimated by ML- and GMM-estimators, respectively.

Farmland Lease Options in the Rural China: Key Determinants and Policy Implications

The study looks at the factors that influence farmland transfer option in Yunnan province, China. To help achieve this objective, 6,007 households were randomly selected from rural communities. The sample includes households who are engaged in lease-out, lease-in, and those who engage in neither lease-out nor lease-in. The data was analysed using multinomial probit model. The results show that farmland certification and knowledge of land tenure policy have positive impact on both lease-in and lease-out decisions relative to the reference category. Farmland size and years of schooling show negative and positive impact on lease-in and lease-out decisions, respectively, relative to the reference category. The net income of the household and crop insurance have positive and negative effect on lease-in and lease-out decisions, respectively, relative to the base category. The results suggest that price per mu has a positive effect on the lease-out decision relative to the base category. The study broadens the scope of analysis on farmland transfer by considering all the options available to a household in farmland transfer decision taking.

Income Diversification by agro- Pastoralists in Cholistan Desert Punjab, Pakistan

The pastoralists are economically depend upon livestock for their income which include their herds of livestock and the bi products produced and sold. The nomads keep moving in search of food and forage so they do not completely destroy the natural resource of a particular area. During their journey of searching water sometimes make them closer to the agriculture land near towns where they used to earn through off farm activities which include the income earned through their unskilled labour activities. The multinomial probit regression employed in this study for analysis revealed that many pastoralists adopt the income diversification strategies which are the coping strategies for other than livestock income to reduce the risks attached with livestock income. The role of Govt. and NGOs for improvement in infrastructure is envisaged to find the enhancement of livestock sector in the area is explored in this study. The study is unique in providing perspective on providing access to different facilities and the role of government in improving living of population.

Strategic entrepreneurial choice between competing crowdfunding platforms

This paper investigates strategic entrepreneurial choice between the UK Big 3 platforms–Crowdcube, Seedrs and SyndicateRoom–that exemplify the three main equity crowdfunding (ECF) shareholder structures identified in the literature. ECF has become a strategic choice for both entrepreneurs and angel and venture capital funds as it offers mutually beneficial advantages to both, especially under the co-investment ECF model where these funds co-invest alongside the crowd. The multinomial probit results show that large founder teams are more likely to choose the co-investment model (SyndicateRoom) but are less likely to opt for the nominee ownership structure (Seedrs). Although less heterogeneous teams are more likely to choose the Seedrs and Crowdcube ownership structures, our results suggest that the probability of choosing the co-investment model (SyndicateRoom) monotonically increases as teams become more heterogeneous. The conclusion is that larger and heterogeneous teams are more likely to raise ECF funds from campaigns explicitly involving professional investors.