|
on Discrete Choice Models |
| By: | Stephan Hetzenecker; Maximilian Osterhaus |
| Abstract: | This paper studies the finite sample performance of the flexible estimation approach of Farrell, Liang, and Misra (2021a), who propose to use deep learning for the estimation of heterogeneous parameters in economic models, in the context of discrete choice models. The approach combines the structure imposed by economic models with the flexibility of deep learning, which assures the interpretebility of results on the one hand, and allows estimating flexible functional forms of observed heterogeneity on the other hand. For inference after the estimation with deep learning, Farrell et al. (2021a) derive an influence function that can be applied to many quantities of interest. We conduct a series of Monte Carlo experiments that investigate the impact of regularization on the proposed estimation and inference procedure in the context of discrete choice models. The results show that the deep learning approach generally leads to precise estimates of the true average parameters and that regular robust standard errors lead to invalid inference results, showing the need for the influence function approach for inference. Without regularization, the influence function approach can lead to substantial bias and large estimated standard errors caused by extreme outliers. Regularization reduces this property and stabilizes the estimation procedure, but at the expense of inducing an additional bias. The bias in combination with decreasing variance associated with increasing regularization leads to the construction of invalid inferential statements in our experiments. Repeated sample splitting, unlike regularization, stabilizes the estimation approach without introducing an additional bias, thereby allowing for the construction of valid inferential statements. |
| Date: | 2024–08 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2408.09560 |
| By: | Meango, Romuald (University of Oxford); Girsberger, Esther Mirjam (University of Technology, Sydney) |
| Abstract: | Ex ante returns, the net value that agents perceive before they take an investment decision, are understood as the main drivers of individual decisions. Hence, their distribution in a population is an important tool for counterfactual analysis and policy evaluation. This paper studies the identification of the population distribution of ex ante returns using stated choice experiments, in the context of binary investment decisions. The environment is characterised by uncertainty about future outcomes, with some uncertainty being resolved over time. In this context, each individual holds a probability distribution over different levels of returns. The paper provides novel, nonparametric identification results for the population distribution of returns, accounting for uncertainty. It complements these with a nonparametric/semiparametric estimation methodology, which is new to the stated-preference literature. Finally, it uses these results to study the preference of high ability students in Côte d'Ivoire for public-sector jobs and how the competition for talent affects the expansion of the private sector. |
| Keywords: | subjective expectations, ex ante returns, nonseparable panel, distribution regression, job search, public sector |
| JEL: | C21 C23 D84 J21 J24 J30 J45 |
| Date: | 2024–07 |
| URL: | https://d.repec.org/n?u=RePEc:iza:izadps:dp17174 |
| By: | Ioannou, Petros; Wang, Zheyu |
| Abstract: | As urban areas grow and city populations expand, traffic congestion has become a significant problem, particularly in regions with substantial truck traffic. This study presents a coordinated freight routing system designed to optimize network utility and reduce congestion through personalized routing guidance and incentive mechanisms. The system customizes incentives and payments for individual drivers based on current traffic conditions and their specific routing preferences. Using a mixed logit model with a linear utility specification, the system captures drivers' route choice behaviors and decisions accurately. Participation is voluntary, ensuring most drivers receive a combined expected utility, including incentives, exceeding their anticipated utility under User Equilibrium (UE). This structure encourages drivers to follow suggested routes. Data collection on drivers' routing choices allows the system to update utility parameter estimates using a hierarchical Bayes estimator, ensuring routing suggestions remain relevant and effective. The system operates over defined intervals, where truck drivers submit their intended Origin-Destination (OD) pairs to a central coordinator. The coordinator assigns routes and payments, optimizing overall system costs and offering tailored incentives to maximize compliance. Experimental results on the Sioux Falls network validate the system's effectiveness, showing significant improvements in the objective function. This study highlights the potential of a coordinated routing system to enhance urban traffic efficiency by dynamically adjusting incentives based on drivers’ choice data and driver behavior. View the NCST Project Webpage |
| Keywords: | Engineering, Social and Behavioral Sciences, Congestion Reduction, Utility Learning, Travel Demand Management, Freight Routing |
| Date: | 2024–08–01 |
| URL: | https://d.repec.org/n?u=RePEc:cdl:itsdav:qt6qb516n9 |
| By: | Buters, Nienke; Prochazka, Jakub; Schefbänker, Katrin; Strobl, Andreas; Teichmann, Karin |
| Abstract: | This report inspects the reproducibility of a study by Dizon-Ross and Jayachandran (2023), which focused on differences in parents' spending on their daughters relative to sons on a large sample of 6, 673 observations in 1, 084 households in Uganda. The original study found that the willingness to pay (WTP) of fathers for different goods for their daughters was lower than for their sons. We were able to computationally reproduce all original results using the original data and code. To test for recreate reproducibility, we tried to reproduce the results of the main analyses using a new code and different software. We were not able to complete the reproduction without analyzing the original code and processed dataset. It was not clear from the manuscript nor the online appendix how the authors dealt with the multilevel structure of the data and how they controlled for different goods, which served as stimulus material. Because the raw data did not have clear labels and the replication package did not include a codebook, we were also unable to identify the variables needed for each analysis. However, after analyzing the original code, we were able to reproduce the original results in MPLUS. The missing code book and missing transcription of survey questions caused complications for investigating robustness reproducibility. Although the authors collected a large number of variables and provided them in the dataset, it was not possible to identify their meaning. Therefore, we were not able to conduct further analyses regarding the main findings of the study. Consequently, we only focused on multicollinearity checks and different constellations of the control variables reported in the paper within the robustness checks. Our analyses showed that the results of the study are robust in this respect. In addition, the missing code book and transcription of survey questions did not allow for direct replicability of the study. Conceptual replicability was not investigated. |
| Date: | 2024 |
| URL: | https://d.repec.org/n?u=RePEc:zbw:i4rdps:137 |
| By: | Andr\'es Aradillas Fern\'andez; Jos\'e Luis Montiel Olea; Chen Qiu; J\"org Stoye; Serdil Tinda |
| Abstract: | We study a class of binary treatment choice problems with partial identification, through the lens of robust (multiple prior) Bayesian analysis. We use a convenient set of prior distributions to derive ex-ante and ex-post robust Bayes decision rules, both for decision makers who can randomize and for decision makers who cannot. Our main messages are as follows: First, ex-ante and ex-post robust Bayes decision rules do not tend to agree in general, whether or not randomized rules are allowed. Second, randomized treatment assignment for some data realizations can be optimal in both ex-ante and, perhaps more surprisingly, ex-post problems. Therefore, it is usually with loss of generality to exclude randomized rules from consideration, even when regret is evaluated ex-post. We apply our results to a stylized problem where a policy maker uses experimental data to choose whether to implement a new policy in a population of interest, but is concerned about the external validity of the experiment at hand (Stoye, 2012); and to the aggregation of data generated by multiple randomized control trials in different sites to make a policy choice in a population for which no experimental data are available (Manski, 2020; Ishihara and Kitagawa, 2021). |
| Date: | 2024–08 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2408.11621 |
| By: | Xiyuan Ren; Joseph Y. J. Chow; Venktesh Pandey; Linfei Yuan |
| Abstract: | As an IT-enabled multi-passenger mobility service, microtransit has the potential to improve accessibility, reduce congestion, and enhance flexibility in transportation options. However, due to its heterogeneous impacts on different communities and population segments, there is a need for better tools in microtransit forecast and revenue management, especially when actual usage data are limited. We propose a novel framework based on an agent-based mixed logit model estimated with microtransit usage data and synthetic trip data. The framework involves estimating a lower-branch mode choice model with synthetic trip data, combining lower-branch parameters with microtransit data to estimate an upper-branch ride pass subscription model, and applying the nested model to evaluate microtransit pricing and subsidy policies. The framework enables further decision-support analysis to consider diverse travel patterns and heterogeneous tastes of the total population. We test the framework in a case study with synthetic trip data from Replica Inc. and microtransit data from Arlington Via. The lower-branch model result in a rho-square value of 0.603 on weekdays and 0.576 on weekends. Predictions made by the upper-branch model closely match the marginal subscription data. In a ride pass pricing policy scenario, we show that a discount in weekly pass (from $25 to $18.9) and monthly pass (from $80 to $71.5) would surprisingly increase total revenue by $102/day. In an event- or place-based subsidy policy scenario, we show that a 100% fare discount would reduce 80 car trips during peak hours at AT&T Stadium, requiring a subsidy of $32, 068/year. |
| Date: | 2024–08 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2408.12577 |