Martina Vandebroek - Academia.edu (original) (raw)
Uploads
Papers by Martina Vandebroek
European Journal of Transport and Infrastructure Research, 2020
Cycling is an important pillar of the global endeavor to have a more sustainable transportation s... more Cycling is an important pillar of the global endeavor to have a more sustainable transportation system. Many papers have studied how trip and person characteristics affect selecting the bike as a transport mode but unlike other researchers, we model the probability of cycling using a binary item response model where the choice is modelled as a trade-off between the individuals' tendency to cycle and the threshold related to each cycling situation. We distinguish between frequent and occasional cyclists. The results show that occasional cyclists are more affected by adverse weather situations, darkness, and uphill slopes. Contrary to the previous studies, a separate bike path turned out a stronger motivator for the group of frequent cyclists. The model fit can substantially be improved by accounting for attribute non-attendance. The results show that weather and wind speed have the highest probability to be taken into account, and the bike path had the lowest probability of being...
The scope of conjoint experiments on which we focus embraces those experiments in which each of t... more The scope of conjoint experiments on which we focus embraces those experiments in which each of the respondents receives a different set of profiles to rate. Carefully designing these experiments involves determining how many and which profiles each respondent has to rate and how many respondents are needed. To that end, the set of profiles offered to a respondent is viewed as a separate block in the design and a respondent effect is incorporated in the model, representing the fact that profile ratings from the same respondent are correlated. Optimal conjoint designs are then obtained by means of an adapted version of the algorithm of Goos and Vandebroek (2004). For various instances, we compute the optimal conjoint designs and provide some practical recommendations.
The Stata Journal: Promoting communications on statistics and Stata
In this article, we describe the randregret command, which implements a variety of random regret ... more In this article, we describe the randregret command, which implements a variety of random regret minimization (RRM) models. The command allows the user to apply the classic RRM model introduced in Chorus (2010, European Journal of Transport and Infrastructure Research 10: 181–196), the generalized RRM model introduced in Chorus (2014, Transportation Research, Part B 68: 224–238), and also the µRRM and pure RRM models, both introduced in van Cranenburgh, Guevara, and Chorus (2015, Transportation Research, Part A 74: 91–109). We illustrate the use of the randregret command by using stated choice data on route preferences. The command offers robust and cluster standarderror correction using analytical expressions of the score functions. It also offers likelihood-ratio tests that can be used to assess the relevance of a given model specification. Finally, users can obtain the predicted probabilities from each model by using the randregretpred command.
International Journal of Electronic Commerce
ABSTRACT In online shopping, e-consumers often choose one among many websites on which to place t... more ABSTRACT In online shopping, e-consumers often choose one among many websites on which to place their orders. The choice depends on key attributes such as trust labels. Presence of such a label shows that the website has been independently certified for online security and privacy. However, consumers may not search for websites with security and privacy seals if they do not know the importance of trust certificates. This behavior of ignoring attributes is called attribute non-attendance. Consumers’ attention to attributes can be increased through provision of information. We investigate the attribute non-attendance switching behavior when information on attributes is provided. Studies have modeled the impact of providing attribute information through changes in preference parameters. We show that an alternative approach is to model the impact via changes in attendance probabilities. We propose that an attribute’s attendance probability post-information depends on its attendance pattern pre-information. Applied to webshop choice data, we find that the proposed model gives a better fit compared with standard approaches. Providing information on attributes led to increases in consumers’ attention to the concerned attributes. Additionally, we found that consumer characteristics affect the shifts in attribute attendance behavior. We show that when assessing effects of providing information, considering the effect on attributes’ attention is important. We provide evidence that availing information on key attributes can give brands a competitive advantage.
Journal of Statistical Software
Discrete choice experiments are widely used in a broad area of research fields to capture the pre... more Discrete choice experiments are widely used in a broad area of research fields to capture the preference structure of respondents. The design of such experiments will determine to a large extent the accuracy with which the preference parameters can be estimated. This paper presents a new R package, called idefix, which enables users to generate optimal designs for discrete choice experiments. Besides Bayesian D-efficient designs for the multinomial logit model, the package includes functions to generate Bayesian adaptive designs which can be used to gather data for the mixed logit model. In addition, the package provides the necessary tools to set up actual surveys and collect empirical data. After data collection, idefix can be used to transform the data into the necessary format in order to use existing estimation software in R.
Journal of Quality Technology
Many production processes consist of successive steps in which things can go wrong without notice... more Many production processes consist of successive steps in which things can go wrong without notice because the problem is only detectable in the final product. For instance in steel manufacturing, the coils undergo melting, hot rolling, annealing and pickling, and defects in one of these stages only become visible after the final process. In other production processes, an output issue may only be detected during final testing after the different parts have been assembled. In all these cases, it is hard to determine which part of the production process is responsible for an unusually high defect rate. We describe a simple procedure based on cluster detection to identify the problematic step if the following conditions are satisfied: the production of defects tends to occur clustered in time and it is feasible to (partially) reorder the part or batch processing sequence in each stage of the production process. Even if reordering is not required for the production, the diagnostic information that can be obtained can well outweigh the potential extra costs involved.
Journal of Quality Technology
When performing an experiment, the observed responses are often influenced by a temporal trend po... more When performing an experiment, the observed responses are often influenced by a temporal trend possibly due to aging of material, learning effects, equipment wear-out, or warm-up effects. The construction of run orders that are optimally balanced for time trend effects usually relies on the incorporation of a parametric representation of the time dependence. Using a parametric approach works very well as long as the unknown time dependence is properly specified or overspecified. However, for complicated temporal trends of unknown periodicity, or when the design size is small compared to the complexity of the response model, a parametric approach may lead to underspecification of the true time trend. Serious problems of bias can result. In this paper we show that, contrary to a fully parametric approach with an underfitted time trend, modeling the time trend nonparametrically is very attractive in terms of both bias and precision of the parameter estimators. An algorithm is presented for the construction of optimal run orders when kernel smoothing is used to model the temporal trend. An industrial example illustrates the practical utility of the proposed design methodology.
European Journal of Transport and Infrastructure Research, 2020
Cycling is an important pillar of the global endeavor to have a more sustainable transportation s... more Cycling is an important pillar of the global endeavor to have a more sustainable transportation system. Many papers have studied how trip and person characteristics affect selecting the bike as a transport mode but unlike other researchers, we model the probability of cycling using a binary item response model where the choice is modelled as a trade-off between the individuals' tendency to cycle and the threshold related to each cycling situation. We distinguish between frequent and occasional cyclists. The results show that occasional cyclists are more affected by adverse weather situations, darkness, and uphill slopes. Contrary to the previous studies, a separate bike path turned out a stronger motivator for the group of frequent cyclists. The model fit can substantially be improved by accounting for attribute non-attendance. The results show that weather and wind speed have the highest probability to be taken into account, and the bike path had the lowest probability of being...
The scope of conjoint experiments on which we focus embraces those experiments in which each of t... more The scope of conjoint experiments on which we focus embraces those experiments in which each of the respondents receives a different set of profiles to rate. Carefully designing these experiments involves determining how many and which profiles each respondent has to rate and how many respondents are needed. To that end, the set of profiles offered to a respondent is viewed as a separate block in the design and a respondent effect is incorporated in the model, representing the fact that profile ratings from the same respondent are correlated. Optimal conjoint designs are then obtained by means of an adapted version of the algorithm of Goos and Vandebroek (2004). For various instances, we compute the optimal conjoint designs and provide some practical recommendations.
The Stata Journal: Promoting communications on statistics and Stata
In this article, we describe the randregret command, which implements a variety of random regret ... more In this article, we describe the randregret command, which implements a variety of random regret minimization (RRM) models. The command allows the user to apply the classic RRM model introduced in Chorus (2010, European Journal of Transport and Infrastructure Research 10: 181–196), the generalized RRM model introduced in Chorus (2014, Transportation Research, Part B 68: 224–238), and also the µRRM and pure RRM models, both introduced in van Cranenburgh, Guevara, and Chorus (2015, Transportation Research, Part A 74: 91–109). We illustrate the use of the randregret command by using stated choice data on route preferences. The command offers robust and cluster standarderror correction using analytical expressions of the score functions. It also offers likelihood-ratio tests that can be used to assess the relevance of a given model specification. Finally, users can obtain the predicted probabilities from each model by using the randregretpred command.
International Journal of Electronic Commerce
ABSTRACT In online shopping, e-consumers often choose one among many websites on which to place t... more ABSTRACT In online shopping, e-consumers often choose one among many websites on which to place their orders. The choice depends on key attributes such as trust labels. Presence of such a label shows that the website has been independently certified for online security and privacy. However, consumers may not search for websites with security and privacy seals if they do not know the importance of trust certificates. This behavior of ignoring attributes is called attribute non-attendance. Consumers’ attention to attributes can be increased through provision of information. We investigate the attribute non-attendance switching behavior when information on attributes is provided. Studies have modeled the impact of providing attribute information through changes in preference parameters. We show that an alternative approach is to model the impact via changes in attendance probabilities. We propose that an attribute’s attendance probability post-information depends on its attendance pattern pre-information. Applied to webshop choice data, we find that the proposed model gives a better fit compared with standard approaches. Providing information on attributes led to increases in consumers’ attention to the concerned attributes. Additionally, we found that consumer characteristics affect the shifts in attribute attendance behavior. We show that when assessing effects of providing information, considering the effect on attributes’ attention is important. We provide evidence that availing information on key attributes can give brands a competitive advantage.
Journal of Statistical Software
Discrete choice experiments are widely used in a broad area of research fields to capture the pre... more Discrete choice experiments are widely used in a broad area of research fields to capture the preference structure of respondents. The design of such experiments will determine to a large extent the accuracy with which the preference parameters can be estimated. This paper presents a new R package, called idefix, which enables users to generate optimal designs for discrete choice experiments. Besides Bayesian D-efficient designs for the multinomial logit model, the package includes functions to generate Bayesian adaptive designs which can be used to gather data for the mixed logit model. In addition, the package provides the necessary tools to set up actual surveys and collect empirical data. After data collection, idefix can be used to transform the data into the necessary format in order to use existing estimation software in R.
Journal of Quality Technology
Many production processes consist of successive steps in which things can go wrong without notice... more Many production processes consist of successive steps in which things can go wrong without notice because the problem is only detectable in the final product. For instance in steel manufacturing, the coils undergo melting, hot rolling, annealing and pickling, and defects in one of these stages only become visible after the final process. In other production processes, an output issue may only be detected during final testing after the different parts have been assembled. In all these cases, it is hard to determine which part of the production process is responsible for an unusually high defect rate. We describe a simple procedure based on cluster detection to identify the problematic step if the following conditions are satisfied: the production of defects tends to occur clustered in time and it is feasible to (partially) reorder the part or batch processing sequence in each stage of the production process. Even if reordering is not required for the production, the diagnostic information that can be obtained can well outweigh the potential extra costs involved.
Journal of Quality Technology
When performing an experiment, the observed responses are often influenced by a temporal trend po... more When performing an experiment, the observed responses are often influenced by a temporal trend possibly due to aging of material, learning effects, equipment wear-out, or warm-up effects. The construction of run orders that are optimally balanced for time trend effects usually relies on the incorporation of a parametric representation of the time dependence. Using a parametric approach works very well as long as the unknown time dependence is properly specified or overspecified. However, for complicated temporal trends of unknown periodicity, or when the design size is small compared to the complexity of the response model, a parametric approach may lead to underspecification of the true time trend. Serious problems of bias can result. In this paper we show that, contrary to a fully parametric approach with an underfitted time trend, modeling the time trend nonparametrically is very attractive in terms of both bias and precision of the parameter estimators. An algorithm is presented for the construction of optimal run orders when kernel smoothing is used to model the temporal trend. An industrial example illustrates the practical utility of the proposed design methodology.