Juan Antonio Carrasco

Assistant Professor

Transportation Section, Department of Civil Engineering, Faculty of Engineering, Universidad de Concepción

Director

Magíster en Ciencias de la Ingeniería, mención Ingeniería Civil, Universidad de Concepción

Telephone (Office): +56 41 220-3603
Fax: +56 41 220-7089
Email: j.carrasco [at] udec [dot] cl
Address:

Department of Civil Engineering,
Faculty of Engineering,
Universidad de Concepción
P.O. Box 160-C,
Concepción, Chile

Resume (adobe acrobat document)

EDUCATION

Doctor of Philosophy, Transport Engineering and Planning

University of Toronto, Department of Civil Engineering, Transportation Engineering and Planning, 2006

Thesis: Social Activity-Travel Behaviour: A Personal Networks Approach

[Eric Pas Dissertation Prize (Honourable Mention), awarded by the International Association for Travel Behaviour Research to the best doctoral dissertations on travel behaviour research in the year 2006]

(Supervisor: Eric J. Miller, Co-supervisor: Barry Wellman)

Master in Engineering Science, Transport Engineering

Pontificia Universidad Católica de Chile, Departamento de Ingeniería de Transporte, Facultad de Ingeniería, 2001

Thesis: ‘Discrete Choice with Correlated Homoskedastic Alternatives: The Nested Logit Model in Depth’

(Supervisor: Juan de Dios Ortúzar)

Industrial Engineer, Diploma in Transport Engineering

Pontificia Universidad Católica de Chile, Departamento de Ingeniería de Transporte, Facultad de Ingeniería, 2001

Bachelor of Engineering Science

Pontificia Universidad Católica de Chile, Departamento de Ingeniería de Transporte, Facultad de Ingeniería, 1999

RESEARCH INTERESTS

Social networks and activity-travel behaviour

Information and communication technologies (ICTs) and travel behaviour

Social exclusion and accessibility issues

Activity-based models and microsimulation approaches

Integrated land use models

Use and estimation of discrete choice and statistical models applied to travel demand analysis

Relationship between urban form and travel decisions

IN THE NEWS

Docentes UdeC: “Juicio a plan de transportes penquista” (El Sur, March 19, 2007). Importance of segregated right of way in public transport as well as the overall need of improving the quality of service of Concepción’s public transport.

PUBLICATIONS

Refereed Publications

Carrasco, J.A. and E. J. Miller, "The social dimension in action: A multilevel, personal networks model of social activity frequency," Transportation Research Part A, forthcoming. Preprint

M. Roorda, M., J.A. Carrasco, and E.J. Miller (2009), “A joint model of vehicle transactions, activity scheduling, and mode choice”, Transportation Research Part B, forthcoming. Abstract

Mok, D., B. Wellman, and J.A. Carrasco, "Does distance matter in the age of the Internet?" Urban Studies, in review process.

Carrasco, J.A., B. Hogan, B. Wellman, and E. J. Miller (2008), "Agency in social activity and ICT interactions: The role of social networks in time and space," Tijdschrift voor Economische en Sociale Geografie (Journal of Economic & Social Geography), forthcoming. Preprint

Carrasco, J.A., E.J. Miller, and B. Wellman (2008), “"How far and with whom do people socialize? Empirical evidence about distance between social network members," Transportation Research Record: Journal of the Transportation Research Board, forthcoming. Preprint

Carrasco, J.A., B. Hogan, B. Wellman, and E. J. Miller (2008), "Collecting social network data to study social activity-travel behaviour: An egocentric approach," Environment and Planning B, 35, forthcoming. Abstract Preprint

Hogan, B., J. A. Carrasco, and B. Wellman (2007), “Visualizing personal networks: Working with participant-aided sociograms”, Field Methods, 19(2) 116-144. Abstract Preprint

Carrasco, J.A. and E. J. Miller (2006), "Exploring the propensity to perform social activities: Social networks approach," Transportation, 33: 463-480. Abstract Preprint

Miller, E.J., M. Roorda, and J.A. Carrasco, (2005), “A Tour-Based Model of Travel Mode Choice”, Transportation, 32: 399-422. Abstract

Carrasco J.A. and J. de D. Ortúzar (2002), “A Review and Assessment of the Nested Logit Model”, Transport Reviews 22(2): 197-218. Abstract

Refereed Book Chapters

Daly, A. and J.A. Carrasco (2008), "The influence of trip length on marginal time and money values," in Proceedings IATBR 2006, R. Kitamura Ed. Elsevier.

Wellman, B., B. Hogan, K. Berg, J. Boase, J. A. Carrasco, R. Côté, J. Kayahara, T. Kennedy, and P. Tran (2006), "Connected Lives: The Project," Chapter 8 in The Networked Neighbourhoods, P. Purcell, Ed. Berlin: Springer. Preprint

Refereed Conference Proceedings and Presentations

Ruminot, N., A. Páez, and J.A. Carrasco (2008) “Vulnerable nodes in networks: An autocorrelation approach,” 48th Congress of the European Regional Science Association, Liverpool, August 27-31, 2008.

Mok, D., B. Wellman and J.A. Carrasco (2008), “Does distance matter in the age of the Internet: Are cities losing their comparative advantage?,” 103^rd Annual Meeting of the American Sociological Association, Boston, August 1-4, 2008. Preprint

Carrasco, J.A., E.J. Miller, and B. Wellman, (2008), "How far and with whom do people socialize? Empirical evidence about distance between social network members," in 87th Annual Meeting of the Transportation Research Board, Washington D.C., January 13-17, 2008. A previous version of this paper was presented at the 53^rd Annual North American Meetings of the Regional Science Association International, Toronto, November 26-28, 2006.

Habib. K.N., J.A. Carrasco, and E.J. Miller (2008), "Social context of activity scheduling: Discrete-continuous model of relationship between "with whom" and episode start time and duration," in 87th Annual Meeting of the Transportation Research Board, Washington D.C., January 13-17, 2008. A previous version of this paper was presented at the XIII Congreso Chileno de Ingeniería de Transporte, Chile, Santiago, October 22-26, 2007.

Carrasco, J.A. (2007), "Redes sociales y comportamiento de transporte: Marco teórico y resultados empíricos", in XIII Congreso Chileno de Ingeniería de Transporte, Chile, Santiago, October 22-26, 2007. (Full text available upon request)

Carrasco, J.A., D. Mok, and B. Wellman (2007), "Integrating distance, travel, and the mode of contact: The Connected Lives Study", Communities and Technologies 2007, Michigan, June 28-30, 2007. (Full text available upon request)

Habib, K.M.N, J.A. Carrasco, and E.J. Miller (2007), "The social dimension of individual’s activity scheduling: How “with whom” influences activity participation", 11^th World Conference on Transport Research, Berkeley, June 24-28, 2007.

Hogan, B., J.A. Carrasco, and B. Wellman (2007), " Maintaining ties near and far: Agency and social accessibility in personal communities", 102^nd Annual Meeting of the American Sociological Association, New York, August 11-14, 2007. A previous version of this paper was presented at the International Sunbelt Social Network Conference XXVII, Corfu Island, Greece, May 1-6, 2007. (Full text available upon request)

Carrasco, J.A. and E. J. Miller (2007), "The social dimension in action: A multilevel, personal networks model of social activity frequency," in 86th Annual Meeting of the Transportation Research Board, Washington D.C., January 21-25, 2007. Preprint

Carrasco, J.A., E.J. Miller, and B. Wellman, (2006), “Spatial and social networks: The case of travel for social activities”, 11th International Conference on Travel Behaviour Research, Kyoto, August 16-20, 2006. Abstract (Full text available upon request)

M. Roorda, M., J.A. Carrasco, and E.J. Miller (2006), “A joint model of vehicle transactions, activity scheduling, and mode choice”, 11th International Conference on Travel Behaviour Research, Kyoto, August 16-20, 2006. Abstract

Daly, A. and J.A. Carrasco (2006), “The influence of trip length on marginal time and money values”, 11th International Conference on Travel Behaviour Research, Kyoto, August 16-20, 2006. Abstract

Carrasco, J.A., B. Hogan, B. Wellman, and E. J. Miller (2006), "Collecting social network data to study social activity-travel behavior: An egocentered approach," in 85th Annual Meeting of the Transportation Research Board, January 22-26, 2006, Washington D.C. [Also presented at MCRI/GEOIDE Second International Colloquium on the Behavioural Foundations of Integrated Land-use and Transportation Models: Frameworks, Models and Applications, Toronto, June 13-14, 2005; and as an invited paper at the workshop Frontiers in Transportation: Social and Spatial Interactions, Amsterdam, Netherlands, July 2-6, 2005]

Carrasco, J.A. and E. J. Miller (2006), "Exploring the propensity to perform social activities: A social networks approach," in 85th Annual Meeting of the Transportation Research Board, January 22-26, 2006, Washington D.C.

Carrasco, J.A. and E. J. Miller (2005), "Socialising with people and not places: Modelling social activities explicitly incorporating social networks," in 9th Computers in Urban Planning and Urban Management Conference, London, June 29 - July 1, 2005. (Full text available upon request)

Miller, E.J., M. Roorda, and J.A. Carrasco, (2003), “A Tour-Based Model of Travel Mode Choice”, 10th International Conference on Travel Behaviour Research, Lucerne, 10-15 August, 2003. Abstract

Carrasco J.A., J de D. Ortúzar, and M. Munizaga (2001), “Una revisión y análisis del modelo Logit Jerárquico”, 8^th Chilean Transport Engineering Conference, 8 -12 October 2001, Concepción (in Spanish). Abstract

Other Selected Relevant Publications, Scholar Research, and Presentations

Carrasco, J.A., S. Ureta and B. Cid (2008), "Actor Network Theory on ICTs and mobility research: An assessment and empirical illustration for social interactions," Third International Specialist Meeting on ICT, Everyday Life and Urban Change. Coombe Lodge, Bristol, UK, March 16-19, 2008.

Mok, D. and J.A. Carrasco (2008), "A comparison of the role of distance in affecting the frequency of contact, pre- and post-Internet", International Sunbelt Social Network Conference XXVIII, Tradewinds Island Resort, St. Pete Beach, Florida, USA, January 22-27, 2008.

Carrasco, J.A., K.M.N. Habib and E.J. Miller (2007), "‘Elicited’ versus ‘revealed’ personal networks: How much social activity-travel can we learn from each?" presented at the workshop Frontiers in Transportation: Social Interactions, Amsterdam, Netherlands, October 14-16, 2007.

Carrasco, J.A., B. Hogan, B. Wellman, and E. J. Miller (2006), "Agency in social activity and ICT interactions: The role of social networks in time and space," invited paper at the Second International Specialist Meeting on ICT. Bergen, Netherlands, November 10-11, 2006. (Full text available upon request)

Carrasco, J.A. (2006), "Social networks: Recent research and future challenges", Double Workshop on Transport and Sustainable Cities. Santiago de Chile, August 7-11, 2006

Carrasco, J.A. (2006), "Interaction of social, physical, and cyberspace: Some evidence from Toronto," invited presentation on the Panel “Cybermobility and Evolution of Personal Travel” at the 85th Annual Meeting of the Transportation Research Board. January 22-26, 2006, Washington D.C.

Daly, A., and J. A. Carrasco (2005), “The influence of trip length on marginal time and money values”, Working Paper, Institute for Transport Studies, University of Leeds.

Wellman, B., K. Berg, J. Boase, J. A. Carrasco, R. Côté, B. Hogan, J. Kayahara, and T. Kennedy (2005), "Connected Lives: How the networked transformation of society affects communication, community and domestic relations," in Internet Generations, 6th International and Interdisciplinary Conference of the Association of Internet Researchers. Chicago, IL, October 5-9, 2005.

Carrasco, J. A. (2005), “Not only how often and where you go, but who you see: social networks and activity-travel behaviour” Presented at the CSpA Seminar Series, March 4, 2005. Centre for Spatial Analysis, McMaster University, Hamilton, ON, Canada. [Also presented as a poster at the MCRI/GEOIDE Second International Colloquium on the Behavioural Foundations of Integrated Land-use and Transportation Models: Frameworks, Models and Applications, Toronto, June 13-14, 2005.]

Hogan, B., J.A. Carrasco, and B. Wellman, (2005), “Generating names in East York: An analog approach”, International Sunbelt Social Network Conference XXV, Redondo Beach, California, February 16-20, 2005.

Carrasco, J. A. (2003) “Social Networks and Activity-travel modelling: Some research ideas” Presented at the MCRI Student/PDF Caucus Workshop, September 14, 2003. University of Toronto, Toronto, ON.

Carrasco, J. A., G. Hutt, P. Pérez, and U. Velarde (2001) “Evaluation of New Public Transport Alternatives Using Stated Preference Techniques”, Internal Research Document, Steer Davies Gleave Consultants, Santiago, Chile (in Spanish).

Carrasco, J. A., A. Daly, and H. Gunn (2000) “Mixed Logit Models to Assess the Value of Time on Destination-Mode Choice Models”, Internal Research Document, The Hague Consulting Group, The Hague, Netherlands.

Carrasco, J. A. (2000) “A Theoretical Revision of the Nested Logit Model”, Working Paper No 66, Department of Transport Engineering, Pontificia Universidad Católica de Chile (in Spanish).

TEACHING

Universidad de Concepción

Modelos de Oferta de Transporte

Introducción a la Ingeniería de Transporte

Planificación de Transporte

Uso de Suelo y Transporte

Métodos de Optimización

University of Toronto

Probability and Statistics for Civil Engineers

ABSTRACTS

A Review and Assessment of the Nested Logit Model

J.A. Carrasco and J. de D. Ortúzar

Abstract

The popular hierarchical or nested logit model that has been the recent source of heated discussions in the literature is reviewed and critically examined. The fundamental underpinnings of the model are first presented in a concise and easy-to-understand fashion and then used to assess each controversy in turn. Monte Carlo simulation is also used to examine some problems that are not possible to address by just resorting to the theory. The main conclusion is that all the model hypotheses are both essential and unambiguous and should not be ignored for a correct use of the model in practice.

A Tour-Based Model of Travel Mode Choice

E.J. Miller, M. Roorda, and J.A. Carrasco

Abstract

This paper presents a new tour-based mode choice model. The model is agent-based: both households and individuals are modelled within an object-oriented, microsimulation framework. The model is household-based in that inter-personal household constraints on vehicle usage are modelled, and the auto passenger mode is modelled as a joint decision between the driver and the passenger(s) to ride-share. Decisions are modelled using a random utility framework. Utility signals are used to communicate preferences among the agents and to make trade-offs among competing demands. Each person is assumed to choose the “best” combination of modes available to execute each tour, subject to auto availability constraints that are determined at the household level. The household’s allocations of resources (i.e., cars to drivers and drivers to ride-sharing passengers) are based on maximizing overall household utility, subject to current household resource levels. The model is activity-based: it is designed to be integrated within a household-based activity scheduling microsimulator. The model is both chain-based and trip-based. It is trip-based in that the ultimate output of the model is a chosen, feasible travel mode for each trip in the simulation. These trip modes are, however, determined through a chain-based analysis. A key organizing principle in the model is that if a car is to be used on a tour, then it must be used for the entire chain, since the car must be returned home at the end of the tour. No such constraint, however, exists with respect to other modes such as walk and transit. The paper presents the full conceptual model and an initial empirical prototype.

A Review and Assessment of the Nested Logit Model

J.A. Carrasco and J. de D. Ortúzar

Abstract

Exploring the propensity to perform social activities: A social networks approach

J.A. Carrasco, and E. J. Miller

Abstract

A conceptual model of social activity-travel behaviour is described, incorporating an activity-scheduling framework which explicitly includes the influence of the individual’s social context. More explicitly, the model develops the concepts of social networks, activities, and social episodes, and defines the individual’s social activity generation and spatial distribution in a context of his/her social networks. Also, empirical findings regarding the influence of social network characteristics on individual’s socialising patterns are discussed. Results suggest that a social networks framework provides interesting insights into the role of physical space, and communication and information technology use. Overall, explicitly incorporating social networks into the activity-travel behaviour modelling framework provides a rich set of insights to understand social activities and the embedded behavioural processes, potentially helping to better understand important issues such as the generation and spatial distribution of activities and travel.

Collecting social network data to study activity-travel behaviour: An egocentred approach

J.A. Carrasco, B. Hogan, B. Wellman, and E. J. Miller

Abstract

Within the study of activity and travel demand, there has been an increasing interest in the effect of social interactions in activity-travel behaviour, that is, how social networks influence activity-travel decisions. Despite this interest, to the authors’ knowledge, no data collection effort has linked explicitly travel behaviour and social interaction, so far. Moreover, although data collection and modelling techniques have gone very far understanding individual activity-travel decision making processes in time and space, we know very little about the linkages between social and spatial interactions. In a companion paper, we showed a conceptual model of social activity-travel behaviour, that links three elements: the characteristics of individuals, their social networks, and the social activity-travel behaviour that arises in this context.

This paper presents the data collection effort designed to partly support that model, including the main research questions and characteristics of the data, the conceptual options taken, the main techniques used, the practical issues and limitations of the method selected, and the expected results.

The data was gathered in the East York area of Toronto between June 2004 and March 2005, as part of the “Connected Lives Study”, a broader study composed by surveys, interviews, and observations about people’s communication patters. The study was conducted by the first author in conjunction with sociologists and social workers of the NetLab, part of the Centre of Urban and Community Studies at the University of Toronto, and lead by the sociologist Barry Wellman. The multidisciplinary setting allowed for a rich crossed-fertilisation in data collection techniques, and a broad set of information gathered.

Having as a motivation the study of social and spatial interactions, this paper discusses the way social network data was collected, and the link made with activity-travel behaviour pattern data. Social network data is traditionally gathered in two ways: as whole and ego-centered networks. Whole network studies consist in asking interviewee about her relationships with closed (pre-defined) list of individuals; however, in an urban setting like the one studied this was unpractical. Then, an ego-centered approach was taken, which consists in asking each interviewee (ego) an open-list of individuals related with her (alters). The way the alter names are elicited heavily determines how many and who are the individuals in the individual’s network; in this case, instead of specific prompting questions, we preferred using a loosely definition of people “very close” and “somewhat close” to the interviewee. Both the ego-centered and the alter eliciting technique have conceptual consequences; the most important is conceiving the personal network as a Ptolemaic universe, with the ego in the centre and the alters around. That is, the data, although collecting interaction among individuals, is still individual-based. Another consequence of open lists is that the relationships between the alter and the ego, and among the alters is left to the interviewee’s criterion.

The paper then focuses in the way social activity-travel behaviour was linked after the social network was built. The interviewee was asked about frequencies of face-to-face and socialising meetings (the last one conceived as visiting, hosting or going to restaurants and pubs) with a sample of alters, and also the frequency of telephone and email interaction, some alter’s characteristics (including the main places they interact with the ego). In addition, we gathered explicit information about time and space characteristics of the last six social events with some of the alters. The main hypothesis behind this way of collecting social travel behaviour data is that the main individual’s social activities arise from the social network, that is, instead of directly start asking about their activities and travel, the main focus is on with whom the individual usually performs social activities. We believe in this way we are able to get a good sense of the individual’s socialising patters, although sacrificing some level of detail.

Finally, we discuss some limitations and the way they were minimised. A first limitation of ego-centered techniques is the risk of prompting a small network. We limited this drawback doing this exercise in the interview stage rather than in the survey, and using an analog instead of a computer approach. Regarding the decision of performing interviews, we believe this helped to gather more data than what we could have gathered in a survey format; controlling for the quality of the network building process; and finally getting a rich set of qualitative data, including comments, extra information, and the interviewee behavioural context. The analog approach we adopted consisted in using paper and pencil rather than a computer interface. This option diminished the burden in the interviewees and – we believe – also increased their motivation, especially because they built in step-by-step procedure their network in a paper, rather than adding names in a computer interface, which usually is a black box for the interviewees.

A second important limitation is the set of sampling procedures from the network. In fact, large networks were usually prompted and a sampling strategy was needed. We developed a sampling strategy based in a combination of the alter’s social closeness attributes and the frequency of interaction. This strategy balanced, on the one hand, capturing a relevant subset of the interviewee’s network (i.e. a relevant sample in their social space), and on the other hand, capturing a relevant set of her activity-travel events (i.e. a relevant sample in their activities and physical space).

The paper also discusses the main qualitative and quantitative information obtained, and how both qualitative and quantitative approaches are complementary and useful to build and inform the conceptual model that motivated this data collection effort. Finally some potential future extensions and methods related with this data collection are discussed.

Visualizing personal networks: Working with participant-aided sociograms

B. Hogan, J.A. Carrasco, and B. Wellman

Abstract

We describe an interview-based data collection procedure for social network analysis designed to (a) aid gathering information about the people known by a respondent and reduce problems with (b) data integrity, and (c) respondent burden. This procedure, a participant-aided network diagram (sociogram), is an extension of traditional name generators. While such a diagram can be produced through computer assisted programs for interviewing (CAPIs) and low-technology (i.e., paper), we demonstrate both practical and methodological reasons for keeping high technology in the lab and low technology in the field. We provide some general heuristics that can reduce the time needed to complete a name generator. We present findings from our Connected Lives field study to illustrate this procedure.

Una revisión y análisis del modelo Logit Jerárquico

J.A. Carrasco, J de D. Ortúzar, and M. Munizaga

Abstract

Sorprende que, tras 25 años de haber sido formulado, aún no se comprendan cabalmente algunas propiedades del modelo logit jerárquico y se discuta su especificación. De hecho, este popular modelo ha sido objeto de una serie de acaloradas discusiones recientemente en la literatura. El objetivo de este trabajo es revisar las principales características del modelo en profundidad y examinarlas críticamente, a fin de aclarar todas las controversias. En primer lugar, se presentan los fundamentos teóricos del modelo, de forma concisa y fácil de entender, para luego examinar cuatro controversias en torno a su especificación: la correcta resolución del problema de no identificabilidad (común a todos los modelos de elección discreta), la aplicabilidad de la popular forma funcional NNNL, utilizada en el paquete estadístico ALOGIT, las reproducción de las particiones de mercado observadas y la pertinencia de una nueva especificación alternativa que provoca confusiones en relación a ciertas propiedades del modelo. Aparte del análisis teórico, se utiliza también simulación de Montecarlo para dilucidar algunos de estos aspectos. En este contexto, la principal conclusión del trabajo es que todas las hipótesis del modelo son esenciales para su correcta especificación e interpretación, y no deben ser ignoradas para un adecuado uso del modelo en la práctica

The Influence of Trip Length on Marginal Time and Money Values

Andrew Daly and J. A. Carrasco

Abstract

The work described in this paper is motivated by the strong empirical finding that the marginal trade-off between time and money in travellers’ decision-making appears to vary with the length of the trip. Specifically, travellers over longer distances appear to have a higher ‘value of time’ (VOT) and this effect appears to be caused by a declining marginal disutility of expenditure, rather than an increasing marginal disutility of spending time. There is no a priori reason to expect that VOT should increase in this way, indeed there seems to be good theoretical reason to expect an opposite trend, so that interest in this issue is natural. In particular, the consequences for project appraisal can be significant: is it correct to attribute a higher value to time savings for travellers making longer journeys than those making shorter journeys?

In practice, analysts have dealt with the effect in different ways for forecasting and appraisal, although the approaches that have been adopted can be justified only on the basis that they are approximations to an unknown ‘true’ underlying model. For example, the use of a ‘log cost’ formulation can improve the fit of models to data but is not reasonable with respect to individuals’ money budgets – the marginal disutility of expenditure must increase as the amount of expenditure increases.

To deal properly with these questions requires that an explanation is given for the observed effect. A number of hypotheses have been suggested to explain the effect and these are reviewed in the paper and their consequences assessed. Several of these hypotheses suggest that the primary mechanism by which VOT increases is through heterogeneity between individuals in taste relative to expenditure. Thus if individuals differ in their sensitivity to price, there will be a self-selection so that those making more expensive journeys will tend to be those with lower sensitivity to price. The effect can be reduced by segmenting the population, for example with respect to income, but an important degree of heterogeneity can remain.

The analysis is set in the Random Utility modelling framework because of the requirement for consistency with economic theory for project appraisal, because this was the basis used for the initial modelling with the data sets concerned and because this framework gives a good basis for modelling.

The work then set out to test the hypothesis of heterogeneity with respect to cost on data sets collected in different ways in different countries: Revealed Preference data from Sydney (1991/2 and 1997/8) and from Paris (1991/2) and Stated Preference data from The Netherlands (1988 and 1997). Existing models which incorporated distance-dependent cost sensitivity terms were reformulated and extended to incorporate heterogeneity of the type described, using a ‘mixed logit’ formulation to handle the heterogeneity. A number of problems had to be solved to reduce the volume of calculation. The analysis was restricted to commuters to keep its scope reasonable.

The results show variation between the data sets but that significant heterogeneity exists in all the data sets analysed. This heterogeneity exists for both time and cost, but more importantly for cost than for time. It appears that, on the whole, the assumption of heterogeneity with a linear cost disutility gives a better explanation of the effects observed – in terms of the fit of the model to the data – than the alternative hypothesis that marginal sensitivity to cost declines with increasing cost. An important further finding is that average VOT and elasticity with respect to time and cost variables can change significantly between different model formulations, and generally are more reasonable when the cost formulation is linear, so that it is quite important to find the right specifications for these models and advantageous to use the formulations with heterogeneity.

The results thus support the hypothesis of self-selection, but also hypotheses that higher prices are not well perceived or are valued proportionately could also be advanced on the basis of the empirical findings. The consequences for of the findings for appraisal would depend on which of these hypotheses were found most plausible. For forecasting, the consequences will remain limited, because it is not yet possible to use mixed logit formulations in large-scale forecasting.

A Joint Model of Vehicle Transactions, Activity Scheduling and Mode Choice

M. Roorda, J. A. Carrasco, and E. J. Miller

Abstract

A joint model of vehicle transactions, activity scheduling and mode choice is estimated based on a retrospective survey of vehicle transactions in the Greater Toronto Area. The vehicle transaction model includes, as explanatory variables, measures of activity/travel stress that are simulated using the Travel Activity Scheduler for Household Agents (TASHA). TASHA is a sophisticated activity-based microsimulation model of activity scheduling and mode choice that represents household interactions of vehicle allocation, ridesharing to joint activities, and drop-off/pick-up of household members. The empirical analysis and model results indicate that there exists an asymmetry in vehicle transactions. Households increase their activity/travel stress far more by disposing of a vehicle, than they alleviate stress by purchasing of a vehicle. Two measures of stress are found to be moderately significant influences on vehicle transactions: the mode choice utility gained by purchasing a vehicle, and the number of conflicts experienced in the household over a limited number of vehicles. This model makes a behavioural contribution by addressing the lack of sophistication typically found in the vehicle usage component of vehicle ownership / transaction models. The model also breaks new ground in the linking of long and short run decisions within a clear conceptual and modelling framework.

Spatial and Social Networks: The Case of Travel for Social Activities

J. A. Carrasco, E. J. Miller, and Barry Wellman

Abstract

Hägerstrand’s argument that Regional Science is about people and not only locations is still a compelling and challenging idea when studying the spatial distribution of activities. In the context of social activity-travel behaviour, this issue is particularly important since the main individuals’ driver to perform a trip is mostly with whom they interact rather than where they go. Using a personal networks approach to measure and study social activity-spaces, which focuses in the relationships of specific individuals (egos) with others (alters), this paper conceptualises social and physical space as a phenomenon beyond the traditional individualistic perspective, showing the need of studying interactions between individuals in a more explicit way. The paper presents the relationships between social activity spaces and the characteristics of egos and their personal networks. Overall, the results strongly suggests that, although –by definition – the spatial distribution of social interaction is an idiosyncratic phenomenon, there are several systematic effects, related with the characteristics of egos and their personal networks, which affect this spatial distribution, and which can serve to better understand where people perform social interactions with others.