Forecasting household demand for light-duty motor vehicles
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
- Resource Relation:
- Other Information: Portions of this document are illegible in microfiche products
- 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
Econometric analysis of the residential demand for electricity in the Pacific Northwest
Related Subjects
29 ENERGY PLANNING
POLICY AND ECONOMY
AUTOMOBILES
DEMAND
FORECASTING
MATHEMATICAL MODELS
DEMOGRAPHY
ECONOMIC ANALYSIS
HOUSEHOLDS
OWNERSHIP
SOCIO-ECONOMIC FACTORS
USA
ECONOMICS
INSTITUTIONAL FACTORS
NORTH AMERICA
VEHICLES
320203* - Energy Conservation
Consumption
& Utilization- Transportation- Land & Roadway
290200 - Energy Planning & Policy- Economics & Sociology