Discrete Choice Modeling

In developing countries, like India, the increasing horizontal spread of the cities and the decentralized nature of development have led to the increased travelling demand. In order to suffice the people’s need for travelling, various transportation modes, having different attributes, have been introduced gradually in the transportation system viz. public city bus, taxi/cab, auto-rickshaw, private vehicle, etc. The trip makers decide the suitability of a mode with respect to his/her criteria which may be classified as quantitative (includes, travel time, travel cost, waiting time etc.) and qualitative (includes, reliability, comfort, convenience, etc.). In this study, the modal share of the Garia-Park Circus corridor in Kolkata, India has been tried to be estimated for work trips only. Transit, Para-transit and Personal vehicles are considered here. The criteria selected for the modeling are Travel Time (TT), Travel Cost (TC), Comfort, Convenience (Waiting Time), Reliability and Dust and Noise. The utility perceived by the trip makers for each of the alternative modes is collected through questionnaire survey and the scaling of the stated preferences is done by psychometric scaling technique.

When making choices, individuals can follow alternative strategies or decision rules to the traditional compensatory utility maximization, raising doubts about to what extent these choices can be used to elicit preferences. In this paper we use a verbal protocol approach to investigate the use of alternative decision rules in discrete choice experiments. The main interest is to identify some of

Self-reported measures are an easy, time-efficient, and low-cost way to collect research data compared to other assessment methods. They are, however, characterized by several limitations regarding the quality and the clarity of the data they produce, especially when they are implemented in education. The main purpose of the current study was to use a method called discrete choice modeling (DCM) in education, in order to develop a self-reported instrument that will reduce the bias for socially desirable responses and to assess teachers' practices in physical activity. DCM method has the ability to overcome the respondents' tendency to answer in a socially desirable way. A nationally representative sample of 531 Greek early educators participated in this study and were administered a self-assessment questionnaire based on DCM. Results showed that the DCM based newly developed instrument manages to assess effectively educators' practices and provided evidence of the applicability of the method in education. Further implications and future recommendations regarding the effective application of DCM in education are discussed.

Choice models are used frequently in marketing to help management make better or more
informed product and pricing decisions.
The most common statistical model for choice studies is the multinomial logit with
heterogeneity via Hierarchical Bayes (HB). The standard approach is to model main effects
for each attribute, but interactions between attributes are also possible. Recently, a company
that produces choice modeling software has introduced an automated interaction search
feature creating additional buzz for interactions in choice studies.
More needs to be known about interactions in the presence of heterogeneity before
applying feature interactions willy-nilly.
Here we propose a method for testing interaction terms with a special look at
brand-price interactions (equivalent to brand-specific price effects) that are often
recommended in practice.
We show that brand-specific pricing is unnecessary when using HB estimation and the
added terms can lead to devastating model overfit.
We provide a general framework for testing interactions. The step-by-step process can
be used without a set of dedicated holdout tasks or sample. When evaluating interactions
and model specifications more generally, overfit as well as managerial inferences should both
be evaluated.

In freight transportation, decisions regarding the choice of transport
mode (or chains of modes) and shipment size are closely linked.
Building on this basic insight, in this paper we estimate and review
various joint econometric models using the Swedish National
Commodity Flow surveys. Robust parameter estimates from this
exercise will be used to update the current deterministic Swedish
national freight model system (the SAMGODS model) to a
stochastic one.

This paper reports on a web-based commercial customer satisfaction study consisting of 1284 respondents, which measured stated attribute importance using six different methods (importance ratings, constant sum, Q-sort, maximum
difference scaling, unbounded ratings and magnitude estimation). Statistical analyses were used to evaluate these six methods in terms of (a) the time they take to administer, (b) their ability to provide discriminating measures and (c) their
predictive validity. Clear winners and losers emerge from these analyses, and applied marketing researchers can use these findings to the benefit of the marketers they support.

Head injury is a major reason for morbidity and mortality worldwide and traumatic head injuries represent the major cause of neurological disability to a clot or hematoma caused by Haemorrhage (ICH) and is the The most common cause of ICH normally reported in our country are road traffic accidents (RTA) followed by falls and assaults. India is a populous country with over a billion every 100,000 population with deprived of these doctors. The unavailability of these specialists is a grave concern to the w care to the nation. The mainstay in the diagnosis of an ICH is the CT (Computed Tomography) scan of the head which is the definitive tool for accurate diagnosis of an ICH following trauma and provides an objective assessment of structural damage to brain. Accurate segmentation of the haemorrhage. This study is on segment Keywords: Intracranial decomposition; Brain haemorrhage segmentation is the first step before detecting the been done on the brain haemorrhage detection using methods like Convolutional neural network other efficient and advanced deep learning techniques. But that is resource intensive. It is also nec efficient when there is a large dataset Hssayeni and colleagues multiple slices and made it public. Second, used deep learning methods to perform segmentation and got a dice coefficient of 31% which is good compared to and colleagues [12] propose entropy based automatic unsupervised brain intracranial haemorrhage segmentation which comprises of FCM clustering, thresholding and edge based active contour methods and they get a better result with the combination than FCM clustering and active use deep learning to diagnose brain haemorrhage. They have used LeNet, GoogleNet and Inception dataset consisting of 100 cases collected from 115 hospitals and discovered LeNet is the among the three. Arjun Majumdar and colleagues haemorrhage instead of Head injury is a major reason for morbidity and mortality worldwide and traumatic head injuries represent the major cause of neurological disability. A traumatic brain injury to a clot or hematoma caused by an accident or any other trauma. (ICH) and is the most common and serious consequence of head injury which can be life The most common cause of ICH normally reported in our country are road traffic accidents (RTA) followed by falls and assaults. India is a populous country with over a billion every 100,000 population with most of them in the urban setup, Indian rural population of more than 70% is deprived of these doctors. The unavailability of these specialists is a grave concern to the w care to the nation. The mainstay in the diagnosis of an ICH is the CT (Computed Tomography) scan of the head which is the definitive tool for accurate diagnosis of an ICH following trauma and provides an objective assessment of ctural damage to brain. Accurate segmentation of the. This study is on segmentation of the brain haemorrhage Intracranial haemorrhage; Discrete wavelet transforms I. RELATED WORK Brain haemorrhage segmentation is the first step before detecting the been done on the brain haemorrhage detection using methods like Convolutional neural network other efficient and advanced deep learning techniques. But that is resource intensive. It is also nec efficient when there is a large dataset, which is not easily available in case of brain haemorrhage. Hssayeni and colleagues [1][2] have contributed in two ways, they collected a new dataset of 82 CT scans with ade it public. Second, used deep learning methods to perform segmentation and got a dice coefficient of 31% which is good compared to other deep learning techniques on small datasets. Indrajeet Kumar propose entropy based automatic unsupervised brain intracranial haemorrhage segmentation which comprises of FCM clustering, thresholding and edge based active contour methods and they get a better result with the combination than FCM clustering and active contour methods alone. use deep learning to diagnose brain haemorrhage. They have used LeNet, GoogleNet and Inception dataset consisting of 100 cases collected from 115 hospitals and discovered LeNet is the among the three. Arjun Majumdar and colleagues [8] haemorrhage instead of traditional methods and achieve a Head injury is a major reason for morbidity and mortality worldwide and traumatic head injuries traumatic brain injury (TBI) is damage to the brain, secondary an accident or any other trauma. This hematoma is known as an Intracranial most common and serious consequence of head injury which can be life The most common cause of ICH normally reported in our country are road traffic accidents (RTA) followed by falls and assaults. India is a populous country with over a billion people and there is approximately one radiologist for of them in the urban setup, Indian rural population of more than 70% is deprived of these doctors. The unavailability of these specialists is a grave concern to the w care to the nation. The mainstay in the diagnosis of an ICH is the CT (Computed Tomography) scan of the head which is the definitive tool for accurate diagnosis of an ICH following trauma and provides an objective assessment of ctural damage to brain. Accurate segmentation of the haemorrhage is the first step before detecting the brain haemorrhage images using discrete wavelet transforms. iscrete wavelet transforms; Segmentation; RELATED WORK Brain haemorrhage segmentation is the first step before detecting the haemorrhage in the brain. A lot of work has been done on the brain haemorrhage detection using methods like Convolutional neural network other efficient and advanced deep learning techniques. But that is resource intensive. It is also nec which is not easily available in case of brain haemorrhage. have contributed in two ways, they collected a new dataset of 82 CT scans with ade it public. Second, used deep learning methods to perform segmentation and got a dice deep learning techniques on small datasets. Indrajeet Kumar propose entropy based automatic unsupervised brain intracranial haemorrhage segmentation which comprises of FCM clustering, thresholding and edge based active contour methods and they get a better result contour methods alone. Tong Duc Phong and colleagues use deep learning to diagnose brain haemorrhage. They have used LeNet, GoogleNet and Inception dataset consisting of 100 cases collected from 115 hospitals and discovered LeNet is the most time [8] use a modified version of U-Net to detect the brain traditional methods and achieve an overall specificity of 98.6% on the small dataset. Brain Haemorrhage Segmentation using Dircrete Wavelet Transform. the terms of the Creative Commons Attribution License; Which Permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source Head injury is a major reason for morbidity and mortality worldwide and traumatic head injuries (TBI) is damage to the brain, secondary This hematoma is known as an Intracranial most common and serious consequence of head injury which can be life-threatening. The most common cause of ICH normally reported in our country are road traffic accidents (RTA) followed by falls people and there is approximately one radiologist for of them in the urban setup, Indian rural population of more than 70% is deprived of these doctors. The unavailability of these specialists is a grave concern to the well-being of the health care to the nation. The mainstay in the diagnosis of an ICH is the CT (Computed Tomography) scan of the head which is the definitive tool for accurate diagnosis of an ICH following trauma and provides an objective assessment of is the first step before detecting the brain images using discrete wavelet transforms. Thresholding; Wavelet haemorrhage in the brain. A lot of work has been done on the brain haemorrhage detection using methods like Convolutional neural network [2][3][5][11] and other efficient and advanced deep learning techniques. But that is resource intensive. It is also necessary and which is not easily available in case of brain haemorrhage. Murtada D. have contributed in two ways, they collected a new dataset of 82 CT scans with ade it public. Second, used deep learning methods to perform segmentation and got a dice deep learning techniques on small datasets. Indrajeet Kumar propose entropy based automatic unsupervised brain intracranial haemorrhage segmentation which comprises of FCM clustering, thresholding and edge based active contour methods and they get a better result Tong Duc Phong and colleagues [13] use deep learning to diagnose brain haemorrhage. They have used LeNet, GoogleNet and Inception-ResNet and a most time-consuming model Net to detect the brain overall specificity of 98.6% on the small dataset.

Empirical studies have shown that demand for multimodal transport systems is highly correlated with activity schedules of individuals. Nonetheless, existing analytical equilibrium models of multimodal systems have only considered trip-based demand. We propose a new market equilibrium model that is sensitive to traveler activity schedules and system capacities. The model is based on a constrained mixed logit model of activity schedule choice, where each schedule in the choice set is generated with a multimodal extension of the Household Activity Pattern Problem. The extension explicitly accounts for both passenger choices of activity participation and multimodal choices like public transit, walking, and vehicle parking. The market equilibrium is achieved with Lagrangian relaxation to determine the optimal dual price of the capacity constraint, and a method of successive averages with column generation finds an efficient choice set of activity schedules to assign flows over the dynamic network load capacities. An example illustrates the model and algorithm, effects similar to Vickrey’s morning commute model can be observed as a special case. A case study of the Oakville Go Transit station access “last mile” problem in the Greater Toronto Area is conducted with 166 survey samples reflecting 3,680 individuals. Results suggest that a $10 fixed parking fee at Oakville station would lead to a reduction of access auto share from 54.8% to 49.5%, an increase in access transit share from 20.7% to 25.9%, and a disutility increase of 11% for the of single-activity residents of Oakville.

Este trabalho apresenta uma análise da escolha de aeroportos por usuários do sistema de transporte aéreo da Região Metropolitana de São Paulo, uma região de múltiplos aeroportos onde é considerada a implantação de um novo aeroporto. Até o presente momento é proposta a utilização do modelo econométrico mixed logit, por ser um modelo mais atual na literatura, robusto e que permite a distribuição aleatória de opções entre os decisores. A base de dados deverá ser coletada com auxílio da Técnica de Preferência Declarada, pois conta com cenário hipotético de um aeroporto inexistente. A análise da competição em uma região de múltiplos aeroportos representa um tema bastante discutido na literatura internacional, onde desenvolveram-se diversos modelos de escolha discreta. No Brasil são poucos os estudos que analisam a competição aeroportuária voltada à percepção dos passageiros. Os modelos são alicerçados na teoria da utilidade e buscam simular o comportamento das pessoas ao escolherem determinado aeroporto para realização de viagens. É importante entender tal comportamento pois o processo de expansão dos aeroportos não depende apenas do aumento da demanda pelo transporte aéreo total, mas de como esta demanda se distribui dentre as alternativas de aeroportos. Em uma região onde existem várias opções de aeroportos, a variação da demanda geral por voos é refletida na demanda de cada aeroporto de maneiras distintas, sendo função de diversos fatores como por exemplo a frequência de voos, o tempo de acesso ao terminal e as tarifas das passagens. Uma das motivações deste estudo é a pressão pela expansão das infraestruturas aeroportuárias que são impostas à aviação civil, devido a uma deficiência no crescimento equivalente de pistas e terminais de passageiros. Com relação à literatura existe ainda uma lacuna na observação da distribuição da demanda brasileira em regiões com mais de um aeroporto. Os resultados deste trabalho poderão contribuir tanto com a comunidade acadêmica, visto que há pouca aplicação na literatura do mixed logit como ferramenta de análise de escolha aeroportuária, como com os administradores aeroportuários e empresas aéreas, na medida em que este assunto é vital para o planejamento de investimentos.

"Kesikli bağımlı değiskeni iki veya daha fazla değer alan modellerde amaç seçimin olasılığının saptanmasıdır. Olasılık değeri bildiği gibi 0 ile 1 arasında bir değer alır. Tercih modellerinin en basiti Doğrusal Olasılık Modelleri’dir. Fakat doğrusal olasılık modellerinde karsılasılan sorunlar yeni tekniklere
gereksinim doğurmustur"  Regresyon, ADF.






























Türkiye ekonometri, Erginbay Uğurlu

The Asia-Pacific region shows potential for growth in the spa market, particularly in Thailand. However, the contrast between the growing day spa industry and the industry's lack of knowledge about day spa users highlights the importance of understanding spa-goers' preferences and considerations with regard to spa selection. Studies of consumers' decision making regarding spa choice commonly cite product quality, therapist quality, ambience quality and price as important determinants. This study used a choice experiment to examine consumers' decisions about day spas via a simulation in which trade-offs were inevitable. The findings suggested that therapist quality was the most dominant attribute in determining spa-goers' choices, followed by product quality, ambience quality, and price. Good therapist and treatment quality led respondents to select spas, whereas poor therapist and treatment quality led them to reject spas. Good ambience quality did not significantly affect spa choice; however, poor ambience quality led to spa rejection. The findings of this study are discussed together with implication for spa managers. The results of this study have important implications for the spa industry in terms of understanding the key drivers of choice among spa customers.

The study examined the impact of Ghana’s fertiliser subsidy programme on maize output and technical efficiency among smallholder farmers in Ghana. Several null hypotheses were tested based on data gathered from a sample of three hundred and fifty-two (352) farmers drawn from two of Ghana’s five agro-ecological zones, the Transition zone and the Guinea savannah zone. Both a probit and stochastic frontier models were used to estimate the determinants of access by farmers to fertiliser subsidy and to assess the impact of the fertiliser subsidy programme on maize output and technical efficiency, respectively. Among others, the study found that age, sex, education, access to improved seed, off farm activities, political influence and distance to the nearest fertiliser retail shop influence farmers’ access to subsidised fertiliser. The study also revealed that access to subsidised fertiliser was low (42.3%) although it positively increases maize output in both Guinea savannah and Transition zones. The study estimated mean TEs of 68.7% and 68.9% in the Transition and the Guinea savannah zones respectively. Also, access to subsidy was found to decrease TE of smallholder farmers in both zones significantly. It is therefore concluded that access to subsidised fertiliser has a mix effect on maize output and needs further investigation.

This study explores how the decision to visit an archaeological museum in Greece is influenced by a range of structural attributes, including admission fee, opening-hours, collection type, and use of multisensory experiences, using the discrete choice model (DCM). DCM is used to study individual preferences towards a set of scenarios that describe a range of choices regarding museum attributes. Study conclusions are drawn from estimated model parameters that determine the most favoured attributes identifying the basis of visiting archaeological museums in Greece. According to the study, participants assess positively the introduction of specific museum services and exhibition features which seem to affect their decision making to visit archaeological museums. The results obtained are useful both from a theoretical point of view, given that there are few writings on this topic, and from a practical point of view, as they provide elements for visitors’ particular needs towards which museum practitioners and cultural policy makers should redirect their attention.

Public policies should reflect and accommodate as much as possible citizens’ preferences and values. However, it is dicult to know the correct value citizens place on public goods as they are not generally exchanged in the marketplace. For this reason,
non market valuation is increasingly considered to be an important tool for informing policy decisions. In this context, Contingent Valuation (CV) and Discrete Choice Experiments (DCE) are probably the most widely used tools to analyse Stated preferences and retrieve people’s Willingness to Pay (WTP) for improving a particular environmental amenity. Since its introduction as a technique of non-market valuation, the number of CV studies published in peer reviewed journals consistently increased for more than three decades. Then, since the mid to late 1990s, the CV technique has been largely substituted by the DCE method.

Having reviewed the literature, it was decided to focus my Phd on DCE method. Notwithstanding the importance of CV as a useful tool for preference elicitation, DCE proved more appealing, as recent literature have identified a number of exciting research
avenues from a methodological and econometric point of view.
Therefore, the PhD has been mainly focused on developing new tools for accommodating heterogeneity in tastes, variances and heuristics employing Latent Class (LC) analysis. As a result, I formalised a class of LC models that can accommodate heteroscedasticity and/or heterogeneity (depending on assumptions and parameterisation) within each class as well as dierent heuristics across classes. The three essays collected
in this thesis represent the main outcome of this research.

More specifically the first paper, on urban land use, introduces class heterogeneity within a LC model. This is obtained by specifying a discrete mixture of sets of continuous distributions. The model is applied to both simulated and real data in order to
demonstrate its flexibility and the advantages for policy appraisal.

The second paper introduces and formalises the idea of a heteroscedastic LC model using data from a recreational site choice study elicited through stated preference methods
to compare various model specifications. Results show that model fit, welfare estimates and choice predictions are sensitive to the manner in which both types of heterogeneity are accommodated for. This is done in WTP-space to directly compare estimates from continuous and discrete mixture representations as well as to demonstrate the importance of including the scale parameter even under this reparameterisation.

In the third and final paper, on species preservation, the well known problem of preference and variance instability due to learning and fatigue in DCE is tackled by applying a scale-adjusted latent class model to uncover both types of instability simultaneously and probabilistically across the sample. Findings highlight the advantages, in terms of model fit, interpretation and policy implications, that can be achieved when both types
of instability are addressed concurrently. Data collected to estimate the existence value of rare and endangered fish species in Ireland are utilised as an empirical case-study.

Finally conclusions and avenues for possible further research and future applications
of the presented models are drawn.

The study was conducted to address the issue of consumers’ willingness to pay for inclusion of cassava flour in bread in Lagos State. The data were collected randomly from a sample of 300 respondents in the state through the use of a well structured questionnaire. The contingent valuation method was adopted to estimate both the mean willingness to pay of consumers and the factors that affect their willingness to pay and these were analyzed using the bivariate probit model. The most significant model was the 10percent cassava flour inclusion as it has the highest number of significant variables (eight) followed by 20percent cassava flour inclusion with seven
significant variables and 30 and 40percent cassava flour inclusion with six significant variables each. The factors that influenced consumers’ willingness to pay for CCW bread were the respondent’s age, gender, respondents’ awareness, married respondents, respondents with head position,and bread share which is the proportion of respondents income spent on bread to total income. The mean willingness to pay for bread with an inclusion of cassava flour at 10, 20, 30, and 40percent cassava flour inclusion were ₦172.70, ₦165.00, ₦154.07, and ₦142.78
respectively for initial bid while the mean willingness-to-pay at 10, 20, 30, and 40percent cassava flour inclusion were ₦180.20, ₦150.41, ₦143.35, and ₦127.36 respectively for subsequent bid.

With the growth in ancillary sales, an area of increasing importance for airlines is the concept of offer management, which entails the creation of dynamic, custom , personalized offers consisting of a flight itinerary and ancillary products offered by an airline. This practice-oriented , overview paper provides an end-to-end, future-oriented framework for determining the composition of optimal base fare and ancillary bundles by customer trip-purpose segment followed by 1:1 personalization to maximize total sales. Our focus in this paper is primarily on the proposed offer management framework and its sub-components.

This paper investigates the role played by different site characteristics in influencing people's choice of outdoor recreation destinations. Contrary to prior studies, our experiment accounts for a large diversity of eligible landscapes described using photomontages. We use a discrete choice experiment (DCE) proposing respondents to choose among hypothetical destinations described in terms of eight site characteristics. We study the trade-offs made by various profiles of respondents among those site characteristics, resulting in different destination choices. The DCE attributes are spatially explicit to represent recreational patterns in the form of site quality maps using Geographic Information Systems (GIS). We conclude by stressing implications of this research for tourism and land management policy-making in peri-urban environments. Interestingly, preferences for site characteristics vary significantly with the recreational activities that respondents engage in. Hikers and cyclists preferences should be particularly considered in future planning decisions.

Firm boundary decisions have been extensively studied using secondary data and surveys investigating the underlying factors and choices made by firms. However, these studies suffer from major limitations as they assume that the boundary choices made by different firms are encompassed in the same set of options and that a singular model operates in the outsourcing domain. In this study, we apply a structured experimental approach to examine the differential effect of potential value creation and value appropriation capability on the boundary choices of firms. We also investigate the degree heterogeneity in outsourcing decision models of managers. Our analysis reveals that managerial preferences for different outsourcing alternatives are more complex than conventional models represent. Further, we find that managers typically apply one of two different cognitive models: (a) one model that focuses predominantly on tangible economic value from cost savings and (b) a second that provides a more balanced assessment of economic and value added rents. Our results show a significant moderating effect of value appropriation capability on the relationship between value creation and boundary choices, which strengthens when potential value creation in an outsourcing relationship involves a shared investment in resources and capabilities.

Young consumers are the most important users of advanced information and communication technologies among others. Particularly, since the gradually increasing smartphone use among young consumers is taken into consideration, an empirical evidence about their purchase behavior in terms of brand perception may contribute to the existing marketing literature. For this purpose, the main objective of this paper is to determine factors that may possibly contribute to young consumers’ smartphone brand preference in a developing country with a particular emphasis on undergraduate students in Turkey. A written-questionnaire is conducted among 1,135 undergraduate students at a well-established university in Turkey. The dependent variable of this study is smartphone brand with a four-category. The data are analyzed using a multinomial logit model due to the unordered nature of the dependent variable. Estimation results suggest monthly individual income and smartphone use in years have an incre...

In this paper, we analyze the travel patterns of Iranian women, where typical patriarchal views and specific social and cultural norms may differ from the patterns of those in western societies. In addition to inherent psycho-physical gender differences, women in Iran can face special constraints forcing them not to be involved in all activity-travel patterns that people in developed countries usually undertake. We pay special attention to the role of marital and employment status on women’s activity-travel patterns. To this end, we develop a joint mode and daily activity pattern (DAP) discrete choice model, which is a two-level mixed nested Logit. The upper nest of the proposed model embodies women’s DAP choices, and the lower nest belongs to the mode choices. In this paper, we try to show how different factors in a patriarchal Muslim society like Iran affect or restrict women’s type and structure of activity-travel patterns.

Abstract Discrete choice models have been intensively used to analyze and predict the behavior of people in transportation modes. In this paper A binary logit model and multinomial logit model have been developed to study a group of policies aimed at discouraging the use of private transportation and this study sought to identify the factors preventing own transport users from shifting to public transport in order to formulate the policies to achieve this.