Modern travel-behavior surveys have become quite complex; they frequently include multiple telephone contacts, travel diaries, and customized stated preference experiments. The complexity and length of these surveys lead to pervasive problems with missing data and non-random response biases. Panel surveys, which are becoming common in transportation research, also suffer from non-random attrition biases. This paper shows how Rubin's (1987a) multiple imputation methodology provides a unified approach to alleviating these problems.

# Your search: "author:Brownstone, David"

## filters applied

## Type of Work

Article (61) Book (0) Theses (5) Multimedia (0)

## Peer Review

Peer-reviewed only (6)

## Supplemental Material

Video (0) Audio (0) Images (0) Zip (0) Other files (0)

## Publication Year

## Campus

UC Berkeley (36) UC Davis (5) UC Irvine (59) UCLA (33) UC Merced (1) UC Riverside (33) UC San Diego (1) UCSF (1) UC Santa Barbara (33) UC Santa Cruz (1) UC Office of the President (27) Lawrence Berkeley National Laboratory (0) UC Agriculture & Natural Resources (0)

## Department

University of California Transportation Center (32) University of California Institute of Transportation Studies (26) Institute of Transportation Studies at UC Berkeley (3) California Partners for Advanced Transportation Technology (1)

UC Davis Institute of Transportation Studies (4) National Center for Sustainable Transportation (4) Plug-In Hybrid & Electric Vehicle Research Center (2)

UC Irvine Institute of Transportation Studies (20)

Department of Economics (3) University of California Energy Institute (1)

## Journal

## Discipline

Social and Behavioral Sciences (29) Engineering (20) Architecture (11)

## Reuse License

BY-NC-ND - Attribution; NonCommercial use; No derivatives (1)

## Scholarly Works (66 results)

This paper discusses important developments in discrete choice modeling for transportation applications. Since there have been a number of excellent recent surveys of the discrete choice literature aimed at transportation applications ( see Bhat, 1997 and 2000a), this paper will concentrate on new developments and areas given less weight in recent surveys. Small and Winston (1999) give an excellent review of the transportation demand literature that includes many examples of how discrete choice models have been used in demand analysis.

Discrete choice modeling is closely related to activity-based modeling of travel demand and duration modeling. Since I have little to add to the excellent recent surveys on these topics by Bhat (2000b) and Bhat and Koppelman (2000), I have restricted this paper to "pure" unordered discrete choice modeling.

The next section discusses recent developments in flexible discrete choice modeling. Note that I define flexible to mean that the parametric model family is rich enough to arbitrarily approximate any discrete choice process consistent with random utility maximization, and I concentrate on the mixed logit model. There is also a relatively new literature which seeks to estimate discrete choice models without making parametric functional form assumptions (see Savin, 2001, Horowitz, 1998, and Koop and Poirier, 2000 for a Bayesian approach). Since this literature is currently limited to binary discrete choice, I have not included it in this paper.

Rubin (1987) has proposed multiple imputations as a general method for estimation ion the presence of missing data. Rubin's results only strictly apply to Bayesian models, but Schenker and Welsh (1988) directly prove the consistency of multiple imputations inferences when there are missing values of the dependent variable in linear regression models. This paper extends and modifies Schenker and Welsh's theorems to give conditions where multiple imputations yield consistent inferences for both ignorable and nonignorable missing data in exogenous variables. One key condition is that the imputed values must have the same conditional first and second moments as the true values. Monte Carlo studies show that the multiple imputation covariance estimates are accurate for realistic sample sizes. They also support the applications of multiple imputations in Brownstone and Valletta (1991), where the multiple imputations estimates substantially changed the qualitative conclusions implied by the model.

This paper discusses important developments in discrete choice modeling for transportation applications. Since there have been a number of excellent recent surveys of the discrete choice literature aimed at transportation applications (see Bhat, 1997 and 2000a), this paper will concentrate on new developments and areas given less weight in recent surveys. Small and Winston (1999) give an excellent review of the transportation demand literature that includes many examples of how discrete choice models have been used in demand analysis.

Modern travel-behavior surveys have become quite complex; they frequently include multiple telephone contacts, travel diaries, and customized stated preference experiments. The complexity and length of these surveys lead to pervasive problems with missing data and non-random response biases. Panel surveys, which are becoming common in transportation research, also suffer from non-random attrition biases. This paper shows how Rubin's (1987a) multiple imputation methodology provides a unified approach to alleviating these problems.

Before discussing solutions to problems caused by missing data and selection, it is important to recognize that their presence causes fundamental problems with identifying models and even "simple" population estimates. Section 2 reviews this work and stresses the need to make generally untestable assumptions in order to carry out any inference with missing data.

We describe and apply choice models, including generalizations of logit called mixed logits, that do not exhibit the restrictive independence from irrelevant alternatives property and can approximate any substitution pattern. The models are estimated on data from a stated-preference survey that elicited customers preferences among gas, electric, methanol, and CNG vehicles with various attributes.

This paper compares results from evaluations of two recent road pricing demonstrations in southern California. These projects provide particularly useful opportunities for measuring commuters' values of time and reliability. Unlike most revealed preference studies of value of time, the choice to pay to use the toll facilities in these demonstrations is relatively independent from other travel choices such as whether to use public transit. Unlike most stated preference studies, the scenarios presented in these surveys are real ones that travelers have faced or know about from media coverage. By combining revealed and stated preference data, some of the studies have obtained enough independent variation in variables to disentangle effects of cost, time, and reliability, while still grounding the results in real behavior. Both sets of studies find that the value of time saved on the morning commute is quite high (between $20 and $40 per hour) when based on revealed behavior, and less than half that amount when based on hypothetical behavior. When satisfactorily identified, reliability is also valued quite highly. There is substantial heterogeneity in these values across the population, but it is difficult to isolate its exact origins. (c) 2004 Elsevier Ltd. All rights reserved.

This chapter forecasts transportation energy demand, for both the U.S. anc California, for the next 20 years. Our guiding principle has been to concentrat~ our efforts on the most important segments of the market. We therefore provide detailed projections for gasoline (58 % of California transportation energy B~in 1988), jet fueI (17%), distillate (diesel) fuel (13%), and residual bunker) fuel (10%). We ignore the remaining 2%--natural gas, aviation gasoIine, liquefied petroleum gas, lubricants, and electricity. Although we discuss prospects for the use of altematlve fuels such as methanoI and natural gas, we do not believe that these will be significant factors in the next 20 years. Table 2-1 gives an overview of transportation energy use in California and the U.S

Our forecasting methodology is based on the principle that predictions should not depend on variables that are themselves difficult to predict; for example, a forecast that uses relative fuel prices as a key component is of little use if it is not possible to determine accurately the relative fuel prices The resulting models are therefore quite simple: they depend only on such factors as demographacs, time trends, and alrplane scrappage patterns 1 Although our proJections do not expicitly model some factors, (e g., the effec of tightened vehicle emission standards, alrcraft noise restrmtIons, fuel prices, and congestion), we do take them into account to the extent that these facto~ were present, anc changing, in data from our modeI-calibratlon periods.

Our predictions are that jet and dmsei fuel demand wili grow at siightly lower than current rates. Gasoline demand wIiI grow at a much slower tare because vebacle ownership is becoming saturated We are unabIe to forecast residuai fuel demand, but it is irrelevant for energy pohcy since there will be a surplus of residual fuel in Califorma for the foreseeable future. Overall, we predict that transportation petroleum demand will grow considerably more slowly than during the last 20 years in both California and the U.S. This suggests that rapid conversion to alternative fuels cannot be justified by demand pressures.