William H Greene

New York University | NYU · Department of Economics (Stern)

Vehicle ownership is an important determinant of the travel demand forecasting process. Vehicle ownership models are used by policy makers to identify factors that affect vehicle miles traveled, and therefore address problems related to energy consumption, air pollution, and traffic congestion. For the conventional travel demand forecasting, it log...

This study presents a new approach to evaluate the economic impact of development projects for cases where baseline and endline data for comparable treated and control samples are available. An important contribution is to bring together propensity score matching (PSM) and difference‐in‐differences (DID) techniques, commonly used in impact evaluati...

Count data with a single mode at zero and a few extreme counts in the outcome variable are frequently seen in medical research. In this article, the response variable—the number of doctor visits in a 3-month interval—exhibited an inflated zero frequency count and a few extreme values. We introduce a right-censored zero-inflated Conway–Maxwell Poiss...

Electricity prices on the European market have decreased significantly over the past few years, resulting in a deterioration of Swiss hydropower firms’ competitiveness and profitability. One option to improve the sector’s competitiveness is to increase cost efficiency. The goal of this study is to quantify the level of persistent and transient cost...

The productive efficiency of a firm can be decomposed into two parts, one persistent and one transient. This distinction seems to be appealing for regulators. During the last decades, public utilities such as water and electricity have witnessed a wave of regulatory reforms aimed at improving efficiency through incentive regulation. Most of these r...

In the aftermath of the Great Recession, a substantial number of families were left financially at risk, especially those with lower income, less wealth, fewer years of education, and unstable employment. This study examined how families responded to the financial stresses of the Great Recession and found that families who were working age, had hig...

Homeownership education and counseling have never been rigorously evaluated through a randomized field experiment. This is a five-year study that addresses this shortcoming. The study is structured to respond to the concerns raised about previous efforts. We employ an . experimental design, with study participants randomly assigned to a control or...

Aims: To examine the effect of transformational leadership on early career nurses' intent to stay, job satisfaction and organizational commitment. Background: Lack of leadership support is one of the top reasons staff nurses leave. Current studies reported mixed results regarding the impact of transformational leadership on key nurse outcomes. H...

We develop a procedure for removing four major specification errors from the usual formulation of binary choice models. The model that results from this procedure is different from the conventional probit and logit models. This difference arises as a direct consequence of our relaxation of the usual assumption that omitted regressors constituting t...

This paper contributes to the literature on the estimation of causal effects by providing an analytical formula for individual specific treatment effects and an empirical methodology that allows us to estimate these effects. We derive the formula from a general model with minimal restrictions, unknown functional form and true unobserved variables s...

Random regret minimization (RRM) interpretations of discrete choices are growing in popularity as a complementary modeling paradigm to random utility maximization (RUM). While behaviorally very appealing in the sense of accommodating the regret of not choosing the best alternative, studies to date suggest that the differences in willingness to pay...

Self-assessed measures of health using Likert-type scales are widely used to assess the health and well-being of populations, and are a feature of household surveys throughout the world. However, the self-reported and subjective nature of these measures means that different people will inherently respond in different ways - a concept known as repor...

In this paper, we examine heterogeneity in the trip preferences of recreationists. We apply a random parameters negative binomial model and a latent class negative binomial model to a panel dataset of beach users at a site on the west coast of Ireland, where access across farmland has been restricted. This approach allows us to account for heteroge...

This paper explores the relationship between different types of relational goods and well-being and examines if there is latent/unobserved heterogeneity in this relationship by applying the latent class ordered probit model. This unique contribution to the literature examines if some individuals have a more associative disposition than others which...

This paper focuses on the self-reported responses given to survey questions of the form "Overall, how would you rate your health?" with typical response items being on a scale ranging from poor to excellent. Usually, the overwhelming majority of responses fall in either the middle category or the one immediately to the "right" of this (for example,...

By measuring twenty streetscape features and numerous other variables for 588 blocks in New York City, we were able to identify variables that explain pedestrian traffic volumes. We found significant positive correlations between three out of twenty streetscape features with pedestrian counts after controlling for density and other built environmen...

We obtain an LM test for the random effects probit model. In the natural parameterization of the model the necessary derivatives are identically zero under the null hypothesis. After a reparameterization, the feasible LM test is based on generalized residuals.

Word of mouth (WOM) is a potent force that can influence the success or failure of new products or services. Its power is amplified and intensified across diverse online and social-media platforms. Such “e-buzz” about movies is a prime example: Can electronic WOM (e-WOM) that takes place prior to a film's release provide an early indication of succ...

This study pools household travel and built environment data from 15 diverse US regions to produce travel models with more external validity than any to date. It uses a large number of consistently defined built environmental variables to predict five household travel outcomes - car trips, walk trips, bike trips, transit trips and vehicle miles tra...

This paper makes two important contributions to the literature on prediction intervals for firm specific inefficiency estimates in cross sectional SFA models. Firstly, the existing intervals in the literature do not correspond to the minimum width intervals and in this paper we discuss how to compute such intervals and how they either include or ex...

We explored direct and indirect influences of physical work environment on job satisfaction in a nationally representative sample of 1,141 early‐career registered nurses. In the fully specified model, physical work environment had a non‐significant direct effect on job satisfaction. The path analysis used to test multiple indirect effects showed th...

As an alternative to sprawling development, smart growth combines proximity to work, proximity to shopping and other destinations, neighborhood housing mix, shared and paid parking, complete street designs, and proximity to public transit. This article uses a stated-choice experiment to determine residents' attitudes toward these various aspects of...

Using multiple choice tasks per respondent in discrete choice experiment studies increases the amount of available information. However, respondents’ learning and fatigue may lead to changes in observed utility function preference (taste) parameters, as well as the variance in its error term (scale); they need to be controlled to avoid potential bi...

We extend the discrete data latent class literature by explicitly defining a latent variable for class membership as a function of both observables and unobservables, thereby allowing the equations defining the class membership and observed outcomes to be correlated. The procedure is then applied to modelling observed obesity outcomes, based upon a...

We discuss the problem of computing the standard errors of functions involving estimated parameters and provide the relevant computer code for three different computational approaches using two popular computer packages. We show how to compute the standard errors of several functions of interest: the predicted value of the dependent variable for a...

Latent class, or finite mixture, modelling has proved a very popular, and relatively easy, way of introducing much-needed heterogeneity into empirical models right across the social sciences. The technique involves (probabilistically) splitting the population into a finite number of (relatively homogeneous) classes, or types. Within each of these,...

The productive efficiency of a firm can be seen as composed of two parts, one persistent and one transient. The received empirical literature on the measurement of productive efficiency has paid relatively little attention to the difference between these two components. Ahn, Good and Sickles (2000) suggested some approaches that pointed in this dir...

Interest in alternative behavioural paradigms to random utility maximisation (RUM) has existed ever since the dominance of the RUM formulation. One alternative is known as random regret minimisation (RRM), which suggests that when choosing between alternatives, decision-makers aim to minimise anticipated regret. Although the idea of regret is not n...

The article first estimates models of mode choice and average trip length for 239 diverse mixed-use developments in six diverse regions. It then applies these models to twelve Leadership in Energy and Environmental Design for Neighborhood Development (LEED-ND) pilot projects to predict approximate vehicle miles of travel (VMT) per person trip. Fina...

Typically, a Poisson regression model is assumed for count data. In many cases, there are many zeros in the dependent variable, therefore the mean is not equal to the variance value of the dependent variable. Thus, we suggest using a hurdle and zero-inflated Poisson regression model. Furthermore, the response variable in such cases is censored for...

In this paper, we develop a panel data negative binomial count model that corrects for endogenous stratification and truncation. We also incorporate a latent class structure into our panel specification, which assumes that the observations are drawn from a finite number of segments, where the distributions differ in the intercept and the coefficien...

In May 2006, the Ministers of Health of all the countries in the African continent, at a special session of the African Union, undertook to institutionalise efficiency monitoring within their respective national health information management systems. The specific objectives of this study were to assess the technical efficiency of National Health Sy...

A Poisson model typically is assumed for count data. In many cases because of many zeros in the response variable, the mean is not equal to the variance value of the dependent variable. Therefore, the Poisson model is no longer suitable for this kind of data. Thus, we suggest using a hurdle negative binomial regression model to overcome the problem...

The proportion of US families that are unbanked (i.e., have no type of checking or savings account) has steadily declined for more than two decades. Nonetheless, more than nine million families still do not participate in the financial mainstream, and roughly half these unbanked families previously held a traditional bank account. This study uses t...

We examine the impact of individual-specific information processing strategies (IPSs) on the inclusion/exclusion of attributes on the parameter estimates and behavioural outputs of models of discrete choice. Current practice assumes that individuals employ a homogenous IPS with regards to how they process attributes of stated choice (SC) experiment...

Two-step estimation with large panel data sets generally involves estimating vectors of individual-specific coefficients in a first-stage. In a second-stage estimation a vector of estimated coefficients is used as the dependent variable. Potential problems of heteroskedasticity in the second stage may be mitigated by weighting all independent obser...

Latent class models offer an alternative perspective to the popular mixed logit form, replacing the continuous distribution with a discrete distribution in which preference heterogeneity is captured by membership of distinct classes of utility description. Within each class, preference homogeneity is usually assumed, although interactions with obse...

This article brings together the stochastic frontier framework with impact evaluation methodology to compare technical efficiency (TE) across treatment and control groups using cross-sectional data associated with the MARENA Program in Honduras. A matched group of beneficiaries and control farmers is determined using propensity score matching techn...

Typically, a Poisson model is assumed for count data. In many cases, there are many zeros in the dependent variable, thus the mean is not equal to the variance value of the dependent variable. Therefore, Poisson model is not suitable anymore for this kind of data because of too many zeros. Thus, we suggest using a hurdle-generalized Poisson regress...

Discrete variables that have an inherent sense of ordering across outcomes are commonly found in large datasets available to many economists, and are often the focus of research. However, assumptions underlying the standard Ordered Probit (which is usually used to analyse such variables) are not always justied by the data. This study provides a rev...

This paper examines the impact of an endogenous cost function variable on the inefficiency estimates generated by stochastic frontier analysis (SFA). The specific variable of interest in this application is endogenous quality in nursing homes. We simulate a dataset based on the characteristics of for-profit nursing homes in California, which we use...

There is a growing interest in traveller behaviour research to explore alternative information processing strategies (often referred to as heuristics or rules) adopted by individuals when assessing packages of attributes describing alternatives in a choice set, and making a choice. One popular attribute processing rule relates to attributes not bei...

A Poisson model typically is assumed for count data. In many cases, there are many zeros in the dependent variable and because of these many zeros, the mean and the variance values of the dependent variable are not the same as before. In fact, the variance value of the dependent variable will be much more than the mean value of the dependent variab...

We make a methodological contribution to the latent class literature by re-examining censored variable analysis within a panel data context. Specifically, we extend the standard latent class tobit panel approach to include random effects, to allow for heteroskedasticity and to incorporate the inverse hyperbolic sine (IHS) transformation of the depe...

It is the purpose of this research to present some evidence on the factors explaining the observed industry differences in incidence and duration of Portuguese strikes. The analysis relies on the proposal of adverse-selection and moral hazard arguments to (partly) explain strikes, interpreting these as costs incurred by more productive workers to s...

The mixed logit model is considered to be the most promising state of the art discrete choice model currently available. Increasingly researchers and practitioners are estimating mixed logit models of various degrees of sophistication with mixtures of revealed preference and stated preference data. It is timely to review progress in model estimatio...

This study explores the relationship between operational efficiency and profitability and growth in the US life Insurance industry, and provides a framework for linking operating performance and financial success. Earnings and growth have particular importance to life insurance companies; earnings and capital determine the viability of the insurer,...

A major concern with the derivation of willingness to pay (WTP) distributions from mixed logit models is the incidence of values over a range that are deemed 'behaviourally questionable', with respect to the sign and magnitude. Recent research in redefining the 'space' within which a choice model is estimated as WTP space, instead of preference-spa...

This paper is a report of a study of factors that affect turnover of newly licensed registered nurses in United States hospitals. There is a large body of research related to nursing retention; however, there is little information specific to newly licensed registered nurse turnover. Incidence rates of turnover among new nurses are unknown because...

In recent years we have seen important extensions of logit models in behavioural research such as incorporation of preference and scale heterogeneity, attribute processing heuristics, and estimation of willingness to pay (WTP) in WTP space. With rare exception, however, a non-linear treatment of the parameter set to allow for behavioural reality, s...

A Poisson model typically is assumed for count data. It is assumed to have the same value for expectation and variance in a Poisson distribution, but most of the time there is over-dispersion in the model. Furthermore, the response variable in such cases is truncated for some outliers or large values. In this paper, a Poisson regression model is in...

This paper investigates factors likely to increase a person's risk of social exclusion, drawing on survey data specifically framed for this purpose. We use a generalised ordered logit model that accounts for observed and unobserved heterogeneity and derive the marginal effects for each influencing attribute. We find that people are less likely to b...

Frontier techniques, including data envelopment analysis (DEA) and stochastic frontier analysis (SFA), have been used to measure health care provider efficiency in hundreds of published studies. Although these methods have the potential to be useful to decision makers, their utility is limited by both methodological questions concerning their appli...

There is a growing literature that promotes the presence of process heterogeneity in the way that individuals evaluate packages of attributes in real or hypothetical markets and make choices. A centerpiece of current research is the identification of rules that individuals invoke when processing information in stated choice (SC) experiments. These...

Latent class models offer an alternative perspective to the popular mixed logit form, replacing the continuous distribution with a discrete distribution in which preference heterogeneity is captured by membership of distinct classes of utility description. Within each class, preference homogeneity is usually assumed (i.e., fixed parameters), althou...

A growing number of empirical studies involve the assessment of influences on a choice amongst ordered discrete alternatives. Ordered logit and probit models are well known, including extensions to accommodate random parameters and heteroscedasticity in unobserved variance. This paper extends the ordered choice random parameter model to permit rand...

There is growing interest in establishing a mechanism to account for scale heterogeneity across individuals (essentially the variance of a variance term or the standard deviation of utility over different choice situations), in addition to the more commonly identified taste heterogeneity in mixed logit models. A number of authors have recently prop...

Regression analysis using panel data for 42 colleges and universities over 14 years suggests that the economics faculty size of universities offering a Ph.D. in economics is determined primarily by the long-run average number of Ph.D. degrees awarded annually; the number of full-time faculty increases at almost a one-for-one pace as the average num...

We examine the interaction effect in nonlinear models discussed by Ai and Norton (2003). Tests about partial effects and interaction terms are not necessarily informative in the context of the model. We suggest ways to examine the effects that do not involve statistical testing.

Plümper and Troeger (2007) propose a three-step procedure for the estimation of a fixed effects (FE) model that, it is claimed, “provides the most reliable estimates under a wide variety of specifications common to real world data.” Their fixed effects vector decomposition (FEVD) estimator is startlingly simple, involving three simple steps, each r...

The Navy’s promotion-retention process involves two successive decisions: The Navy decides whether an individual is selected for promotion, and then, conditional on the Navy’s decision, the sailor decides whether to reenlist or leave the Navy. Rates of promotion and retention depend on individuals’ demographic and other characteristics, wars...

To examine the relationship between registered nurse (RN) staffing mix and quality of nursing home care measured by regulatory violations. A retrospective panel data study (1999-2003) of 2 groups of California freestanding nursing homes. One group was 201 nursing homes that consistently met the state's minimum standard for total nurse staffing leve...

Nursing turnover is costly for health care organizations. Newly licensed registered nurses work behavior is a complex process, influenced by their attitudes toward their work, personal characteristics, job opportunities, and workplace attributes. Several characteristics are significant in predicting satisfaction (ethnicity, gender) and organization...

There have been profound changes in both political and economic institutions in China over the last 20 years. Moreover, the pace of transition has led to variation across the country in the level of development. In this paper, we use panel data for the Chinese provinces to study the role of legal institutions, financial deepening and political plur...

We present several extensions of the Poisson and negative binomial models for count data based on the lognormal model for latent heterogeneity in the conditional means. The lognormal model provides a versatile specification that is more flexible than the familiar log gamma form, and provides a platform for several “two part” extensions that have ap...

It is increasingly common for analysts to seek out the opinions of individuals and organizations using attitudinal scales such as degree of satisfaction or importance attached to an issue. Examples include levels of obesity, seriousness of a health condition, attitudes towards service levels, opinions on products, voting intentions, and the degree...

This paper focuses on inequality in the distribution of household health resources across children for three developing transition countries; Azerbaijan, the Kyrgyz Republic and Tajikistan. In particular, we analyse the household's decision to pursue preventative care for their children. The econometric specification utilized is a random effects pr...

Random effects estimates using panel data for 42 colleges and universities over 16 years reveal that the economics faculty size of universities offering a Ph.D. in economics is determined primarily by the long-run average number of Ph.D. degrees awarded annually; the number of full-time faculty increases at almost a one-for-one pace as the average...

We detail the basic theory for models of discrete choice. This encompasses methods of estimation and analysis of models with discrete dependent variables. Entry level theory is presented for the practitioner. We then describe a few of the recent, frontier developments in theory and practice.

This paper focuses on how the distribution of household health resources can be influenced by bargaining within the household with a particular focus on gender biases across children for four countries; Armenia, Azerbaijan, Kyrgyzstan and Tajikistan. The framework utilised is owed to McElroy and Horney (1981). Following McElroy and Horney the value...

This note develops an encompassing model for two well known variants of the negative binomial model (the NB1 and NB2 forms). We conclude with an application of the proposed model using the data employed in a recent health care study.

Obesity is a major risk factor for several diseases including diabetes, heart disease and stroke. Increasing rates of obesity internationally are set to cost health systems increasing resources. In the US a conservative estimate puts resources already spent on obesity at $120 billion annually. Given scarce health care resources it is important that...

Many studies have examined predictors of nurses' intention to work in their job, including desire to quit. Intent has been a good predictor of actual turnover. Few longitudinal studies exist that consider regional variables. To extend the conceptual framework of turnover research to the whole nursing workforce and determine: (1) how do demographics...

It has become popular practice that joint estimation of choice models that use stated preference (SP) and revealed preference (RP) data requires a way of adjusting for scale to ensure that parameter estimates across data sets are not confounded by differences in scale. The nested logit ‘trick’ presented by Hensher and Bradley in 1993 [Hensher, D.A....

This essay examines the role of data and program-code archives in making economic research 'replicable.' Replication of published results is recognized as an essential part of the scientific method. Yet, historically, both the 'demand for' and 'supply of' replicable results in economics has been minimal. 'Respect for the scientific method' is not s...

This chapter provides an overview of techniques for econometric analysis of technical (production) and economic (cost) efficiency. The stochastic frontier model of Aigner, Lovell, and Schmidt (1977) is now the standard econometric platform for this type of analysis. The underlying models and econometric techniques that have been used in studying te...

We survey the literature on models for ordered choices, including ordered logit and probit specifications. The contemporary form of the model is presented and analyzed in detail. The historical development of the model is presented as well. We detail a number of generalizations that have appeared in the recent literature. Finally, we propose a new...

This paper examines how the institutional set-up of public research organisations (PROs) affects how firms are able to utilise direct interaction with publicly employed researchers. We argue that the role that PRO interaction has to play in the firm’s innovation processes depend on the organisational and cognitive distances between the firm and the...

Count response regression models refer to regression models having a count as the response; e.g., hospital length of stay, number of bacterial pneumonia cases per zip code in Arizona from 2000 to 2005. Poisson regression is the basic model of this class. Having an assumption of the equality of the distributional mean and variance, Poisson models ar...

Developments in simulation methods, and the computational power that is now available, have enabled open-form discrete choice models such as mixed logit to be estimated with relative ease. The random parameter (RP) form has been used to identify preference heterogeneity, which can be mapped to specific individuals through re-parameterisation of the...

This paper examines the information content of insider trade Form 4 filings under the more timely disclosure regime introduced by Section 403 of the Sarbanes-Oxley Act of 2002 (SOX). Abnormal returns and trading volumes around filings of insider purchases are significantly greater after than before SOX. The increase in returns around post-SOX filin...

This paper introduces a four-state failure model to depict a wider range of distress scenarios that public companies typically face in the real world. We use a multinomial error component logit model to analyse firm failure, a major advance on the modelling techniques used in previous research. The error component logit model, being an extension of...

This study presents several extensions of the most familiar models for count data, the Poisson and negative binomial models. We develop an encompassing model for two well known variants of the negative binomial model (the NB1 and NB2 forms). We then propose some alternative approaches to the standard log gamma model for introducing heterogeneity in...

We consider a bivariate Poisson model that is based on the lognormal heterogeneity model. Two recent applications have used this model. We suggest that the correlation estimated in their model frameworks is an ambiguous measure of the correlation of the variables of interest, and may substantially overstate it. We conclude with a detailed applicati...

The most familiar fixed effects (FE) and random effects (RE) panel data treatments for count data were proposed by Hausman, Hall and Griliches (HHG) (1984). The Poisson FE model is particularly simple and is one of a small few known models in which the incidental parameters problem is, in fact, not a problem. The same is not true of the negative bi...

I modify the uniform-price auction rules in allowing the seller to ration bidders. This allows me to provide a strategic foundation for underpricing when the seller has an interest in ownership dispersion. Moreover, many of the so-called "collusive-seeming" equilibria disappear.