Forecasting household demand for light-duty motor vehicles
Technical Report
·
OSTI ID:6245667
This report describes an operational nationwide motor vehicle demand forecasting system based on a set of empirically estimated disaggregated discrete choice models. The forecasting system represents household behavior in terms of two interrelated decisions: the number of motor vehicles to own (ownership level); and, given the ownership level decision, the specific vehicle types that are chosen. Separate models are estimated to explain vehicle type choice conditional on each ownership level. These models are each multinomial logit specifications, where the probability of choosing any given vehicle (or combination of vehicles) is a function of the utility of the chosen vehicles relative to the sum of the utilities of all feasible vehicle alternatives. Each household has the choice of maintaining its existing vehicle or replacing it with any other new or used car or light-duty truck available on the market. The utility of a given vehicle is a function of the household's travel requirements (as expressed by household size, number of workers, annual vehicle use, etc.) and vehicle attributes (price, fuel economy, roominess, etc.). These relationships are expressed in a linear-in-parameters utility function in which the empirically determined parameters express the relative importance households associate with each specific vehicle attribute. The vehicle ownership level decision is also represented by a multinomial logit specification. The models are designed to explain the vehicle type choice behavior of individual households as a static holdings process. Specifically, each household is viewed as evaluating its vehicle holdings at a specific point of time and adjusting its choices so as to maximize utility. Forecasting with the motor vehicle demand models is accomplished by summing the vehicle type choice predictions over a demographically weighted sample of US households.
- Research Organization:
- Booz, Allen and Hamilton, Inc., Bethesda, MD (USA)
- DOE Contract Number:
- AC05-84OR21400
- OSTI ID:
- 6245667
- Report Number(s):
- DOE/OR/21400-T243; ON: DE86007474
- Country of Publication:
- United States
- Language:
- English
Similar Records
A personal vehicle transactions choice model for use in forecasting demand for future alternative-fuel vehicles
A Machine-Learning Decision-Support Tool for Travel-Demand Modeling
What Makes You Hold on to That Old Car? Joint Insights From Machine Learning and Multinomial Logit on Vehicle-Level Transaction Decisions
Technical Report
·
Fri Dec 30 23:00:00 EST 1994
·
OSTI ID:419140
A Machine-Learning Decision-Support Tool for Travel-Demand Modeling
Conference
·
Thu Feb 07 23:00:00 EST 2019
·
OSTI ID:1494741
What Makes You Hold on to That Old Car? Joint Insights From Machine Learning and Multinomial Logit on Vehicle-Level Transaction Decisions
Journal Article
·
Sun Jul 03 20:00:00 EDT 2022
· Frontiers in Future Transportation
·
OSTI ID:1874736
Related Subjects
29 ENERGY PLANNING, POLICY, AND ECONOMY
290200 -- Energy Planning & Policy-- Economics & Sociology
32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATION
320203* -- Energy Conservation
Consumption
& Utilization-- Transportation-- Land & Roadway
AUTOMOBILES
DEMAND
DEMOGRAPHY
ECONOMIC ANALYSIS
ECONOMICS
FORECASTING
HOUSEHOLDS
INSTITUTIONAL FACTORS
MATHEMATICAL MODELS
NORTH AMERICA
OWNERSHIP
SOCIO-ECONOMIC FACTORS
USA
VEHICLES
290200 -- Energy Planning & Policy-- Economics & Sociology
32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATION
320203* -- Energy Conservation
Consumption
& Utilization-- Transportation-- Land & Roadway
AUTOMOBILES
DEMAND
DEMOGRAPHY
ECONOMIC ANALYSIS
ECONOMICS
FORECASTING
HOUSEHOLDS
INSTITUTIONAL FACTORS
MATHEMATICAL MODELS
NORTH AMERICA
OWNERSHIP
SOCIO-ECONOMIC FACTORS
USA
VEHICLES