Demand Model Research Articles

Evaluating the Impacts of Parameter Uncertainty in a Practical Transportation Demand Model

The inherent uncertainty in travel forecasting models—arising from potential and unknown errors in input data, parameter estimation, or model formulation—is receiving increasing attention from both the scholarly and practicing communities. In this research, we investigate the variance in forecasted traffic volumes resulting from varying the mode and destination choice parameters in an advanced trip-based travel demand model. Using Latin hypercube sampling to construct several hundred combinations of parameters across the plausible parameter space, we introduce substantial changes to implied travel impedances and modal utilities, on the order of a 10 percent variation. However, the aggregate effects of these changes on forecasted traffic volumes are small, with a variation of approximately 1 percent on high-volume facilities. It is likely that in this example—and perhaps in others—the network assignment places constraints on the possible volume solutions and limits the practical impacts of parameter uncertainty. Nevertheless, parameter uncertainty may not be the largest contributor to error in practical travel forecasts. Further research should examine the robustness of this finding across other less constrained networks and within activity-based travel model frameworks.

A Decision-Making Tool for Sustainable Energy Planning and Retrofitting in Danish Communities and Districts

This study presents a novel framework for city-level energy planning and retrofitting, tailored to Danish cities and neighborhoods. The framework addresses the challenges of large-scale urban energy modeling by integrating automated processes for data collection, energy demand prediction, and renewable energy integration. It combines open-source simulation tools and validated datasets, enabling efficient and scalable predictions of energy performance across urban areas, including streets, districts, and entire cities, with minimal user input. The key components include data collection and demand modeling, energy resource estimation, performance gap evaluation, and the design of retrofitting strategies with renewable energy integration. The DanCTPlan energy planning tool, developed based on this framework, was applied to two case studies in Denmark: a single street with 101 buildings and a district comprising five streets with 1284 buildings. In the single-street case, retrofitting all buildings to meet current regulations resulted in a 60.8% reduction in heat demand and a 5.8% reduction in electricity demand, with significant decreases in peak energy demands. The district-level retrofitting measures led to a 29.5% reduction in heat demand and a 2.4% reduction in electricity demand. Renewable energy scenarios demonstrated that photovoltaic systems supplying 30% of electricity demand and solar thermal systems meeting 10% of heating demand would require capacities of 2218 kW and 3540 kW, respectively. The framework’s predictive capabilities and flexibility position it as a robust tool to support decision-makers in developing sustainable and cost-effective energy strategies, paving the way toward establishing energy-efficient and positive energy districts.

Estimating Models of Supply and Demand: Instruments and Covariance Restrictions

We consider the identification of empirical models of supply and demand with imperfect competition. We show that a restriction on the covariance between unobserved demand and cost shocks can resolve endogeneity and identify the price parameter. We demonstrate how to employ this approach in estimation, and we compare it to the method of instrumental variables. Our formal results also indicate that weaker covariance restrictions can bound the price parameter. We illustrate the covariance restriction approach with applications to ready-to-eat cereal, cement, and airlines. (JEL C51, D12, L13, L61, L66, L93)

A multi-output deep learning model for energy demand and port availability forecasting in EV charging infrastructure

A multi-output deep learning model for energy demand and port availability forecasting in EV charging infrastructure

How stable and predictable are welfare estimates using recreation demand models?

How stable and predictable are welfare estimates using recreation demand models?

Heterogeneous Preferences for Neighborhood Amenities: Evidence from GPS Data

Abstract How do preferences for neighborhood amenities vary by income? Using data on over 100 million visits to 1.4 million establishments, I estimate a discrete choice model of demand for restaurants, shops, personal services, and entertainment places. While preferences for specific establishments often vary by income, preferences for neighborhoods' overall amenity access are highly aligned. The primary determinant is density: dense urban areas have sufficient variety to offer broad appeal. For incumbents of gentrifying neighborhoods, tailoring amenities to higher-income entrants has only modest welfare effects relative to the effects of potential displacement to cheaper neighborhoods with worse access to amenities.

Amplitude-Scaling Bias Analysis of Ground Motion Record Set in Strip Method for Structural Seismic Fragility Assessment

The multi-strip method is often used to establish a demand model for fragility analysis. Using the multi-strip method to scale the ground motion record may cause uncertainty and bias in structural response calculation and fragility assessment. It is necessary to analyze the effect of differences in the amplitude scaling range in different strips on structural seismic response calculation and seismic fragility assessment. In this paper, the multi-strip method was used to analyze the seismic demand bias based on four multi-story reinforced concrete frame structures subjected to eight ground motion record sets. The bias, variance, and coefficient of variation in different strips in each group of ground motion records were obtained. The effect of different strips on the demand bias was investigated by analysis of variance (ANOVA). Uncertainty quantification of structural demand and fragility curves was carried out using the bootstrap sampling method. The results for structures in different ground motion record sets verify that the differences between the demand bias for different strips by amplitude scaling are statistically insignificant for a 95% confidence level. These findings will contribute to the use of scaling methods for ground motion record sets in a probabilistic seismic demand assessment, allowing for a more reliable prediction of structural seismic fragility.

Individual differences in working memory and attentional control continue to predict memory performance despite extensive learning.

Individual differences in working memory predict a wide range of cognitive abilities. However, little research has been done on whether working memory continues to predict task performance after repetitive learning. Here, we tested whether working memory ability continued to predict long-term memory (LTM) performance for picture sequences even after participants showed massive learning. In Experiments 1-3, subjects performed a source memory task in which they were presented a sequence of 30 objects shown in one of four quadrants and then were tested on each item's position. We repeated this procedure for five times in Experiment 1 and 12 times in Experiments 2 and 3. Interestingly, we discovered that individual differences in working memory continually predicted LTM accuracy across all repetitions. In Experiment 4, we replicated the stable working memory demands with word pairs. In Experiment 5, we generalized the stable working memory demands model to attentional control abilities. Together, these results suggest that people, instead of relying less on working memory, optimized their working memory and attentional control throughout learning. (PsycInfo Database Record (c) 2025 APA, all rights reserved).

Demand forecasting in clearance season for fast fashion retail: a residual learning approach

PurposeThis study focuses on forecasting demand for markdown pricing in situations where historical data are limited and proposes using a hybrid knowledge-based residual (KRL) network to enhance the accuracy of demand predictions in fast fashion retailing.Design/methodology/approachWe use a hybrid KRL network structure to increase forecasting accuracy in fast fashion data by combining artificial neural network (ANN) and theoretical demand models (TDMs) such as linear, exponential and multinomial logit. We used a linear demand model (LDM) as a theoretical demand model, which is one of the popular TDMs in the literature, and combined it with neural networks utilizing the residual network structure. We tested it on real fast fashion data to estimate the next week’s demand at the clearance seasons of 5 years.FindingsThe results underscore KRL’s capability to derive mutual benefits from both neural networks and LDM, especially in the specific context of limited fast fashion data. Furthermore, KRL outperforms LDM and ANN models when used individually in forecasting accuracy.Originality/valueThe research paper proposes a scientific, quantitative method for forecasting the sales for markdown settings, combining data driven and TDMs using residual learning thereby resolving the issue of insufficient sales data in the fashion industry.

ELECTRICITY DEMAND FORECASTING IN MALAYSIA USING SEASONAL BOX-JENKINS MODEL

The development of a precise forecasting model for electricity demand is essential for optimizing the efficiency of planning within the power generation sector. The electricity demand data in Malaysia exhibits seasonal patterns, making it necessary to evaluate the forecasting capabilities of the Box-Jenkins model for predicting weekly peak electricity demand. The objective of this study is to assess how well the Box-Jenkins model performs in forecasting the weekly peak electricity demand. This study utilizes weekly electricity demand data, specifically the highest values recorded each week, measured in megawatts (MW), spanning from 2005 to 2016. The findings indicate that SARIMA (4,1,0)(0,1,0)52 is the best-suited choice for predicting electricity demand. This conclusion is supported by its notably low values of Root Mean Square Error (RMSE), Mean Absolute Error (MAE) and Mean Absolute Percentage Error (MAPE) which stand at 623.3015, 488.5673, and 2.95%, respectively. The MAPE value of the suggested model, falling below the 5% threshold, suggests that the seasonal Box-Jenkins model performs quite effectively when it comes to predicting electricity demand in the context of Malaysian data. To summarize, the proposed seasonal Box-Jenkins model exhibits significant potential and delivers promising performance when forecasting electricity demand characterized by seasonal patterns.

Hotelling Meets Wright: Spatial Sorting and Measurement Error in Recreation Demand Models

Hotelling Meets Wright: Spatial Sorting and Measurement Error in Recreation Demand Models

Revenue-Maximizing Shared Parking and Electric Vehicle Charging Management in Multi-Unit Dwellings

In urban areas, searching for parking and electric vehicle (EV) charging can result in cruising, congestion, and environmental externalities. Recognizing the business opportunity of offering private parking and charging infrastructure access within multi-unit dwellings (MUDs) during daytime, we model a shared parking and EV charging management system. We maximize the revenue of MUD charging hubs in mixed land use, catering to public demand. Our approach accounts for the objectives of the two stakeholders involved: a demand model is fitted on the choices of EV charging users, and the supply model optimizes the allocation of parking and charging requests in an MUD parking lot. A binary integer linear programming model for the allocation of parking and charging spaces with a rolling horizon is integrated with matching rules that handle both parking and charging requests. In our numerical experiments in a neighborhood of Chicago, Illinois, we estimate the performance of the MUD parking and charging system with metrics that include revenue, number of matchings, and utilization rates. At any given time, MUDs with lower prices attract more charging requests, particularly those of longer duration, resulting in higher revenue and greater charging utilization. Dynamic pricing facilitates a more equitable distribution of requests; as MUD parking lots reach capacity and their fees increase, other MUDs become more competitive, attracting additional requests. Comparing our method against first-come-first-served and optimal-solution benchmarks, we demonstrate our model’s effectiveness in dynamically managing mixed parking and charging demand in MUD charging hubs.

Online Learning and Pricing for Multiple Products With Reference Price Effects

ABSTRACTWe consider the dynamic pricing problem of a monopolist seller who sells a set of mutually substitutable products over a finite time horizon. Customer demand is sensitive to the price of each individual product and the reference price which is formed from a comparison among the prices of all products. To maximize the total expected profit, the seller needs to determine the selling price of each product and also select a reference product (to be displayed) that affects the consumer's reference price. However, the seller initially knows neither the demand function nor the optimal reference product, but can learn them from past observations on the fly. As such, the seller faces the classical trade‐off between exploration (learning the demand function and reference price) and exploitation (using what has been learned thus far to maximize revenue). We propose a rate‐optimal dynamic learning‐and‐pricing algorithm that integrates iterative least squares estimation and bandit control techniques in a seamless fashion. We show that the cumulative regret, that is, the expected revenue loss caused by not using the optimal policy over periods, is upper bounded by where hides any logarithmic factors. We also establish the regret lower bound (for any learning policies) to be . We then generalize our analysis to a more general demand model. Our algorithm performs consistently well numerically, outperforming an exploration‐exploitation benchmark. The use of price experimentation and estimation techniques could be readily applied in real retail management.

The impact of long-term care insurance on the health status and healthcare expenditure of older adults in China.

Long-term care needs have grown with population aging. This study explores the relationships among health status, healthcare expenditure, and long-term care insurance (LTCI) among the older adults. Using data from the China Health and Retirement Longitudinal Study (CHARLS 2011, 2013, 2015, 2018, and 2020) and based on the demand model for healthcare services, this study employed the difference-in-difference (DID) method to assess how the implementation of LTCI contributed to the health status and healthcare expenditures of the older adults. The propensity score matching DID method (PSM-DID) and other tests were used to conduct further robustness checks. The findings demonstrate a significant positive impact of LTCI on improving the health status and reducing healthcare expenditures in the elderly population. The PSM-DID indicates that LTCI can effectively improve the health status of the elderly population and reduce their healthcare spending. Based on the findings, the development of relevant policy frameworks for LTCI in China is recommended. These include consider the differences among the various pilot cities and social strata to allow policy adjustments and improvements in a timely, establish a dynamic and diversified long-term care insurance financing mechanism, encourage collaboration between medical institutions and elder care facilities, establish effective contact between LTCI and medical institutions, and use incentive policies such as tax relief to provide financial support and subsidies.

Forecasting tourist arrivals using STL-XGBoost method

Forecasting tourism demand in a timely manner is critical for ensuring the smooth operation of the tourism industry. Over time, time series models have been widely applied to estimate the number of tourists arriving. In this paper, we proposed a XGBoost model for tourism demand forecasting based on the STL seasonal decomposition. The first phase of our proposed model involves applying STL decomposition to preprocess the time series, separating it into two components: the seasonal and de-seasonal terms. During the second phase, the seasonal term is modeled and predicted with the Holt-Winters model. For the de-seasonal term, the ARIMA model is first employed to capture the residual part, Then, the XGBoost model is utilized to reconstruct both the de-seasonal term and its lag, along with the residual part obtained from the ARIMA model. By integrating the forecast outputs from both the Holt-Winters and XGBoost models, the final tourism demand predictions can be derived. The effectiveness of the proposed model is demonstrated using the tourist arrivals data in Macau from eight countries: United States, Germany, Malaysia, Philippines, India, Thailand, Italy and Korea (South Korea). The validation results indicate that the proposed model exhibits superior forecasting performance for time series data showing seasonality and trendency, simultaneously enhancing interpretability without increasing model complexity. The model outperforms five benchmark comparison models when assessed using the Symmetric Mean Absolute Percent Error (SMAPE), Root Mean Square Error (RMSE), Mean Absolute Error (MAE), and Mean Absolute Percentage Error (MAPE) metrics.

A Game Theory Approach to Trip Distribution Model: Game Distribution Model

Introduction In Transport planning, the four-step travel demand model is one of the most popular macro scale planning tools since the 1970s. The second step of this traditional model, i.e., trip distribution, plays an essential role in the model and at the same time, it is the most controversial step. Methods Various alternative approaches are available in the literature for modelling trip distribution, such as the Gravity Model, intervening opportunities model, logit models etc. However, none of these models is universally accepted. Although the gravity model is the most common and well-known trip distribution model, it is often subject to the same criticisms as other alternative models. In this study, the trip distribution step of the four-step model is designed as a rational game to replicate preferences in a more realistic way. The usual assumption in the traditional distribution model is that decision-makers are influenced by only origin/destination-based travel attributes and/or impedance of the trip to the destination. However, other probable destination circumstances are not included in this procedure. In order to incorporate the actual determinants of trip making, Traffic Analysis Zones (TAZ) are considered to be gamers in the proposed model, and the generalized trip cost was considered to be the cost of gamers. Each TAZ has a utility that attracts individuals from other TAZs. Trips are distributed by comparing the utility of a TAZ with the associated cost.. Empirical verification of the model is performed by using the household survey data for Eskişehir/Turkey. Results The evaluation was performed with different goodness of fit statistics for 5 different structured O/D matrices representing various demand settings. The results indicate that the proposed model demonstrates strong performance according to selected micro- and macro-level goodness-of-fit statistics, with all R2 values exceeding 0.80. Conclusion When compared to the classic Gravity Model, these goodness-of-fit measures yield better results in general.

Spatio-temporal forecasting model for EV charging demands

Spatio-temporal forecasting model for EV charging demands

Development of A Demand Model for School Trips in Colombo, Sri Lanka

Development of A Demand Model for School Trips in Colombo, Sri Lanka

Inference and Impact of Category Captaincy

This paper studies category captaincy, a vertical relationship where retailers delegate shelf placement and pricing decisions of an entire product category to one of the leading manufacturers within that category. These contracts involve preferential treatment given to certain brands within the retailer. However, they are confidential, and empirical analysis has been limited. To study the prevalence and welfare impact of these contracts, I develop a novel approach to infer the presence of captaincy contracts based on preferential treatment in shelf placement and pricing. I first classify retailers into different captaincy types based on a brand × retailer specific quality of shelf placement inferred from a demand model. I then conduct pricing tests and find that captains, in line with theoretical predictions and industry anecdotes, eliminate double markups from their own products but not from competitor products. Comparative statics show that captaincy relationships increase the market share of the captain, but they can also increase consumer welfare by about 5% due to the elimination of double markups on the captain’s products. This paper was accepted by Raphael Thomadsen, marketing. Supplemental Material: The online appendices and data files are available at https://doi.org/10.1287/mnsc.2023.02039 .

Tax policy and tobacco consumption in Botswana: an ARDL-EC approach

PurposeBotswana imposed a 30% ad valorem tobacco tax in 2014 to reduce tobacco use. The purpose of the paper is to assess the effectiveness of the tax in curtailing tobacco consumption.Design/methodology/approachAn autoregressive distributed lag and equilibrium correction (ARDL-EC) framework is applied on data for the period 1975–2020 to estimate a dynamic tobacco demand model. The estimated price elasticity is then used to quantify the effects of the tax on tobacco consumption.FindingsA 10% rise in the tobacco price results in a 6.6% decrease in tobacco consumption, suggesting an inelastic response. A 10% rise in income yields a 12% increase in tobacco consumption, reflecting that the rapid economic growth Botswana experienced post-independence yielded increased tobacco use. Tobacco consumption declined by 3.6% per year, possibly capturing the effects of increasing awareness of the adverse effects of tobacco use over time. The 30% tobacco tax yielded a 20% reduction in tobacco consumption, suggesting moderate effectiveness in curtailing consumption. The tax reduced annual tobacco consumption by 100 grams per capita or 151 metric tons nationally.Research limitations/implicationsFuture research could explore the effects of non-price anti-tobacco measures and socioeconomic and demographic factors on tobacco use to provide further insights for guiding the development of targeted anti-smoking interventions.Originality/valueTobacco demand elasticities vary across countries and analytical methods. Therefore, country-specific empirical evidence is essential for policymaking. An existing study in Botswana employed cross-sectional analysis, which does not capture the addictive effects of tobacco. The ARDL-EC framework is employed to close this gap. Simulated effects of the tax are useful for policy reform in Botswana.Peer reviewThe peer review history for this article is available at: https://publons.com/publon/10.1108/IJSE-01-2024-0097