|
on Discrete Choice Models |
| By: | Theile, Philipp (Energiewirtschaftliches Institut an der Universitaet zu Koeln (EWI)) |
| Abstract: | With the growing adoption of electric vehicles, understanding user charging behavior is increasingly important for informing operational, investment, and policy decisions regarding their integration into the power system. While utility functions are commonly used to describe user preferences in charging behavior models, most existing studies rely on formulations with limited theoretical consistency and empirical validation, potentially leading to biased expectations. This paper empirically compares different utility function specifications and examines their implications for charging behavior modeling and charging station profitability. I introduce a novel discrete choice model framework to efficiently estimate utility function parameters from revealed preference data. Using a dataset of observed charging sessions at public charging stations in Germany, the model identifies accurate utility functions, uncovers charging preferences, and simulates station segment viability. The results suggest that charging utility is non-linear: marginal utility decreases with charged energy and marginal disutility increases with charging duration. An interaction between energy and duration leads to higher marginal valuation of energy for longer charging durations. Stations profit from inelastic demand driven by users who highly value energy content, are less price sensitive, and engage in high-value activities at the charging location, such as in urban areas or traffic hubs. |
| Keywords: | Electric Vehicles; charging behavior; utility function; discrete choice model; revealed preference data; charging station viability |
| JEL: | C25 C44 C53 Q40 R40 |
| Date: | 2025–07–02 |
| URL: | https://d.repec.org/n?u=RePEc:ris:ewikln:2025_007 |
| By: | Christoph Engel (Max Planck Institute for Research on Collective Goods, Bonn) |
| Abstract: | In an experiment on the large language model GPT-4o, a supplier always makes a higher profit if it replaces uniform contract terms with a set of terms between which the custom-er may choose. The extra profit results from price discrimination. There is a first order and a second order effect. The first order effect results from heterogeneous willingness to pay for a more protective term. The second order effect results from the possibility that con-tract choice is a signal for general willingness to pay for the traded commodity. In the ex-periment, the effect is bigger if the least protective version is labelled as the default, and more protective terms as an “upgrade†. The effect is smaller if, conversely, the most pro-tective version is labelled as the default and less protective (and cheaper) versions as an opportunity for “savings†. The effect is also bigger if the supplier only sets the price after it knows which version of the contract the consumer chooses. The profit increasing effect of giving the consumer a choice is strong. Most pieces of demographic information (which the supplier might, for instance, learn from cookie data) have a significantly smaller effect on profit. If the supplier combines cookie information about demographic markers with contract choice, it often even makes an extra profit. The main results replicate on Gemini 2.5 flash. |
| Date: | 2025–06 |
| URL: | https://d.repec.org/n?u=RePEc:mpg:wpaper:2025_08 |
| By: | Venkatachalam Anbumozhi (Economic Research Institute for ASEAN and East Asia (ERIA)); Kaliappa Kalirajanl (Emeritus Professor, Australian National University); Ayu Pratiwi Muyasyaroh (Economic Research Institute for ASEAN and East Asia (ERIA)); Veerapandian Karthick (Assistant Professor, Institute for Social and Economic Change, Bengaluru, India) |
| Abstract: | Carbon pricing is a policy tool designed to account for the external costs of carbon emissions, such as damage to crops, healthcare costs, and property loss due to climate change. It attaches a price to these costs and allocates responsibility to the sources of emissions. This approach helps incentivise the reduction of carbon emissions and encourages the adoption of technologies aimed at achieving a net zero economy. Revenue generated from carbon pricing can be reinvested by companies to support sustainable practices, including employee benefits and health insurance. While a few countries in the Association of Southeast Asian Nations (ASEAN) and East Asia have implemented carbon pricing mechanisms, there is limited understanding of individual preferences regarding these mechanisms at the national and regional levels. The Carbon Border Adjustment Mechanism of the European Union aims to standardise carbon prices for internationally traded products. However, there is a lack of knowledge about preferences for such policy instruments across key stakeholders and countries. A survey has been conducted to elicit stakeholders’ preferences and willingness to pay (WTP) for a carbon price in ASEAN and East Asia. The overall proportion of ‘yes’ answers to the WTP question was around 70%. Mean WTP corresponds to an additional price of US$10–US$15. The analysis of more than 500 consumer responses revealed that several modifiers impact the choice of higher and lower WTP additional costs for climate actions. Amongst the consumer groups, academia and household residents are more concerned about climate change and its harmful consequences but have less knowledge and lower appreciation of external pressures such as the European Union’s Carbon Border Adjustment Mechanism. This, coupled with the already high electricity price, could have resulted in the lower WTP by the private sector respondents. Three null hypotheses on the effects of WTP on carbon emission reductions, revenue recycling, and regional cooperation are tested. The low WTP underscores the urgency of measures to overcome market size and technical and financing constraints, and to address regulatory hurdles that raise transaction costs, to achieve industrial competitiveness |
| Keywords: | carbon price; climate change; net zero economy; revenue recycling; willingness to pay; ASEAN and East Asia |
| JEL: | Q49 Q58 C46 |
| Date: | 2025–06–04 |
| URL: | https://d.repec.org/n?u=RePEc:era:wpaper:dp-2025-03 |
| By: | Marcin P\k{e}ski; Colin Stewart |
| Abstract: | A researcher wants to ask a decision-maker about a belief related to a choice the decision-maker made; examples include eliciting confidence or cognitive uncertainty. When can the researcher provide incentives for the decision-maker to report her belief truthfully without distorting her choice? We identify necessary and sufficient conditions for nondistortionary elicitation and fully characterize all incentivizable questions in three canonical classes of problems. For these problems, we show how to elicit beliefs using variants of the Becker-DeGroot-Marschak mechanism. |
| Date: | 2025–06 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2506.12167 |
| By: | Daniele Caliari (Università di Napoli Federico II, CSEF and UCFS, Uppsala University); Valentino Dardanoni (Università di Palermo); Carla Guerriero (Università di Napoli Federico II and CSEF); Paola Manzini (School of Economics, University of Bristol); Marco Mariotti (School of Economics and Finance, Queen Mary University of London) |
| Abstract: | Choice mistakes may be consequential. While we have plentiful evidence on adult behaviour, childrenÕs choices are much less studied, yet not only may they shed light on adult behaviour, but they are themselves important, as potentially leading to low educational attainment, unhealthy food choices, and risky behaviours. In this paper, we study experimentally how childrenÕs choice consistency and ability to avoid mistakes change with age. We study choice by primary school children in two (ubiquitous) domains: riskless and risky choice. We elicit complete choice functions over deterministic choices, while for lotteries we introduce a novel experimental design, documenting as a particular type of framing effect, consistent with correlation neglect, so far only studied in adults. With plentiful evidence of choice errors in adults, unsurprisingly choice errors and inconsistencies abound in children - strikingly though, in some cases already by age 10-11 children display error rates which are close to those observed in adults. Our results are well captured by a model of limited, stochastic consideration. Our experiment is rich enough to highlight the shape that potential interventions could take, aiming at increasing childrenÕs consideration capacity. Different socioeconomic backgrounds seem to matter, though, reassuringly, the gap does tend to close over time. |
| Keywords: | correlation neglect, bounded rationality, violations of first order stochastic dominance. |
| JEL: | D01 D90 |
| Date: | 2024–11–15 |
| URL: | https://d.repec.org/n?u=RePEc:sef:csefwp:737 |
| By: | Chlond, Bettina; Goeschl, Timo; Lohse, Johannes |
| Abstract: | Randomized controlled trials remain underutilized in informing policy design, despite their potential. Moral objections to experimentation (“experiment aversion”) have been proposed as an explanation. We present three studies with members of the general public and policy-makers that allow us to measure and compare moral approval, stated preferences as well as revealed preferences for policy experimentation, within the overarching context of a public assistance program. We find that evidence based on moral approval systematically underestimates revealed preferences for policy experimentation due to conceptual misalignment and hypothetical bias. People and policy-makers trade off possible moral objections against the benefits of policy experimentation. |
| Date: | 2025–07–01 |
| URL: | https://d.repec.org/n?u=RePEc:awi:wpaper:0763 |
| By: | Ryota IWAMOTO; Takunori ISHIHARA; Takanori IDA |
| Abstract: | This study empirically investigates the differences in risk preferences and loss aversion between humans and generative AI. We conduct a nationwide online survey of 4, 838 individuals and generate AI responses under identical conditions by using personas constructed from demographic attributes. The results show that in gain domains, both humans and the AI select risk-averse options and exhibit similar preference patterns. However, in loss domains, AI shows a stronger risk-loving tendency and responds more sharply to individual attributes such as gender, age, and income. We retrain the AI by fine-tuning it based on human choice data. After fine-tuning, the AI’s preference distribution moves closer to that of humans, with loss-related decisions showing the greatest improvement. Using Wasserstein distance, we also confirm that fine-tuning reduces the behavioral gap between AI and humans. |
| Keywords: | bias, bias, loss aversion, risk preference, generative AI, persona, fine-tuning, Wasserstein distance |
| JEL: | D91 C91 |
| Date: | 2025–07 |
| URL: | https://d.repec.org/n?u=RePEc:kue:epaper:e-25-006 |
| By: | Hoogland, Kelly PhD; Hardman, Scott PhD |
| Abstract: | Battery-electric vehicles (BEVs) are central to California’s strategy to reduce transportation-related emissions; however, low-income households face significant structural barriers to adoption. These barriers include the high upfront purchase costs of new BEVs, limited supply of used BEVs, limited access to home charging, and low awareness of BEVs. To better understand these obstacles and identify effective policy responses, our research team analyzed survey data collected from 2, 051 priority population households throughout California between December 2023 and June 2024. The survey asked households about their vehicle purchasing behavior, ownership costs, and socio-demographics. |
| Keywords: | Engineering |
| Date: | 2025–06–01 |
| URL: | https://d.repec.org/n?u=RePEc:cdl:itsdav:qt5996v4gn |
| By: | Alda Botelho Azevedo (Instituto de Ciências Sociais, Universidade de Lisboa); Inês Gonçalves (Nova School of Business and Economics, Universidade NOVA de Lisboa); João Pereira dos Santos (Queen Mary University of London, ISEG – University of Lisbon, and IZA) |
| Abstract: | Our study investigates public opinion on the housing affordability crisis in Portugal through a nationally representative survey combined with an information provision experiment. Participants were asked to identify perceived causes of rising housing prices, assess their factual knowledge of the housing market and sociodemographic trends, and indicate their preferred policy solutions, carefully framed to reflect trade-offs. Half of the respondents were randomly assigned to receive official statistical information on these trends before indicating their policy preferences. The findings reveal significant heterogeneity in beliefs about the causes of the crisis, pervasive misperceptions regarding market trends, and a limited impact of information provision on policy preferences. These results underscore the challenges of addressing housing policy through informational interventions alone and highlight the need for strategies that integrate behavioral and contextual factors to foster informed public engagement. |
| Keywords: | Real estate prices; Information-provision experiment; Population; Tourism; Portugal |
| JEL: | R31 F60 J18 |
| Date: | 2025–06 |
| URL: | https://d.repec.org/n?u=RePEc:mde:wpaper:191 |
| By: | Monteiro, Sofia; Pujol-Busquets, Georgina; Smith, James; Larmuth, Kate |
| Abstract: | There is reasonable concern that self-reported nutrition assessments do not reflect actual food choices. Yet, a correspondence between both is imperative to evaluate any intervention on food preferences. This paper makes such a comparison. It provides evidence from a low-carbohydrate nutrition education program, which is assessed with both surveys and an incentivized behavioral measure of food choice. The main result is that there is a large correspondence between survey and behavioral measures for our sample of 95 women from two historically underprivileged communities in the Western Cape, South Africa. Compared to the control, the treatment group reported a 35% lower intake from the high-carbohydrate/ ultra-processed food Red List and 60% higher intake from the low-carbohydrate whole foods Green List. The treatment group was also 40% less likely to buy anything from the Red List with a supermarket voucher. In terms of the Green List, the treatment group was significantly more likely to buy eggs, organ meat, traditional fats, avocado and fish but there was no difference in red meat and chicken, non-starchy vegetables and full cream dairy. Low-cost incentivized measures of revealed preferences can be designed to validate subjective habits, increasing confidence in the quality of evidence from nutrition intervention studies. |
| Date: | 2025 |
| URL: | https://d.repec.org/n?u=RePEc:zbw:ifwkie:319531 |
| By: | Charles Shaw |
| Abstract: | The double/debiased machine learning (DML) framework has become a cornerstone of modern causal inference, allowing researchers to utilise flexible machine learning models for the estimation of nuisance functions without introducing first-order bias into the final parameter estimate. However, the choice of machine learning model for the nuisance functions is often treated as a minor implementation detail. In this paper, we argue that this choice can have a profound impact on the substantive conclusions of the analysis. We demonstrate this by presenting and comparing two distinct Distributional Instrumental Variable Local Average Treatment Effect (D-IV-LATE) estimators. The first estimator leverages standard machine learning models like Random Forests for nuisance function estimation, while the second is a novel estimator employing Kolmogorov-Arnold Networks (KANs). We establish the asymptotic properties of these estimators and evaluate their performance through Monte Carlo simulations. An empirical application analysing the distributional effects of 401(k) participation on net financial assets reveals that the choice of machine learning model for nuisance functions can significantly alter substantive conclusions, with the KAN-based estimator suggesting more complex treatment effect heterogeneity. These findings underscore a critical "caveat emptor". The selection of nuisance function estimators is not a mere implementation detail. Instead, it is a pivotal choice that can profoundly impact research outcomes in causal inference. |
| Date: | 2025–06 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2506.12765 |
| By: | Yulia Evsyukova (University of Mannheim and ZEWÐLeibniz Centre for European Economic Research); Federico Innocenti (Università di Verona); Niccolò Lomys (CSEF and Università degli Studi di Napoli Federico II) |
| Abstract: | We study how framing interplays with information design. Whereas Sender conceives all contingencies separately, Receiver cannot initially distinguish among some of them, i.e., has a coarse frame. To influence Receiver’s behavior, Sender first decides whether to refine Receiver’s frame and then designs an information structure for the chosen frame. Sender faces a trade-off between keeping Receiver under the coarse frame — thus concealing part of the information structure — and reframing — hence inducing Receiver to revise preferences and prior beliefs after telling apart initially indistinguishable contingencies. Sender benefits from re-framing if this enhances persuasion possibilities or makes persuasion unnecessary. Compared to classical information design, Receiver’s frame becomes more critical than preferences and prior beliefs in shaping the optimal information structure. Although a coarse worldview may open the doors to Receiver’s exploitation, re-framing can harm Receiver in practice, thus questioning the scope of disclosure policies. |
| Keywords: | Framing; Information Design; Disclosure Policies. |
| JEL: | D1 D8 D9 G2 G4 M3 |
| Date: | 2024–12–01 |
| URL: | https://d.repec.org/n?u=RePEc:sef:csefwp:743 |
| By: | Jin Seo Cho (Yonsei University) |
| Abstract: | The current study investigates testing the mixture hypothesis of Poisson regression models using the likelihood ratio (LR) test. The motivation of the mixture hypothesis stems from the unobserved heterogeneity, and the null hypothesis of interest is that there is no unobserved heterogeneity in the data. Due to the nonstandard conditions described in the text, the LR test does not weakly converge to the standard chi-squared random variable under the null hypothesis. We derive its null limit distribution as a functional of the Hermite Gaussian process. Furthermore, we introduce a methodology to obtain the asymptotic critical values consistently. Finally, we conduct Monte Carlo experiments and compare the power of the LR test with the specification test developed by Lee (1986). |
| Keywords: | Mixture of Poisson Regression Models; Likelihood Ratio Test; Asymptotic Null Distribution; Gaussian Process. |
| JEL: | C12 C22 C32 C52 |
| Date: | 2025–07 |
| URL: | https://d.repec.org/n?u=RePEc:yon:wpaper:2025rwp-254 |
| By: | Yasumasa Matsuda; Rei Iwafuchi |
| Abstract: | This paper proposes a novel framework for modeling time series of probability density functions by extending auto-regressive moving average(ARMA) models to density-valued data. The method is based on a transformation approach, wherein each density function on a compact domain [0, 1]d is approximated by a B-spline mixture representation. Through generalized logit and softmax mappings, the space of density functions is transformed into an unconstrained Euclidean space, enabling the application of classical time series techniques. We define ARMA-type dynamics in the transformed space. Estimation is carried out via least squares for density-valued AR models and Whittle likelihood for ARMA models, with asymptotic normality derived under the joint divergence of the time horizon and basis dimension. The proposed methodology is applied to spatio-temporal human population data in Tokyo, where meaningful temporal structures in the distributional dynamics are successfully captured. |
| Date: | 2025–06–23 |
| URL: | https://d.repec.org/n?u=RePEc:toh:dssraa:146 |
| By: | Yao Luo; Peijun Sang |
| Abstract: | We propose a class of sieve-based efficient estimators for structural models (SEES), which approximate the solution using a linear combination of basis functions and impose equilibrium conditions as a penalty to determine the best-fitting coefficients. Our estimators circumvent repeated solution of the structural model, apply to a broad class of models, and are consistent, asymptotically normal, and asymptotically efficient. Moreover, they solve unconstrained optimization problems with fewer unknowns and offer convenient standard error calculations. As an illustration, we apply our method to an entry game between Walmart and Kmart. |
| Keywords: | Efficient Estimation, Sieves, Empirical Games, Joint Algorithm, Nested Algorithm |
| JEL: | C51 C57 C45 |
| Date: | 2025–06–22 |
| URL: | https://d.repec.org/n?u=RePEc:tor:tecipa:tecipa-801 |