Search results

In this paper we use Monte Carlo sampling experiments to examine the properties of pretest estimators in the random parameters logit (RPL) model. The pretests are for the presence of random parameters. We study the Lagrange multiplier (LM), likelihood ratio (LR), and Wald tests, using conditional logit as the restricted model. The LM test is the fastest test to implement among these three test procedures since it only uses restricted, conditional logit, estimates. However, the LM-based pretest estimator has poor risk properties. The ratio of LM-based pretest estimator root mean squared error (RMSE) to the random parameters logit model estimator RMSE diverges from one with increases in the standard deviation of the parameter distribution. The LR and Wald tests exhibit properties of consistent tests, with the power approaching one as the specification error increases, so that the pretest estimator is consistent. We explore the power of these three tests for the random parameters by calculating the empirical percentile values, size, and rejection rates of the test statistics. We find the power of LR and Wald tests decreases with increases in the mean of the coefficient distribution. The LM test has the weakest power for presence of the random coefficient in the RPL model.

Interest in car-sharing initiatives, as a tool for improving transport network efficiency in urban areas and on interurban links, has grown in recent years. They have often been proposed as a more cost effective alternative to other modal shift and congestion relief initiatives, such as public transport or highway improvement schemes; however, with little implementation in practice, practitioners have only limited evidence for assessing their likely impacts.

This study reports the findings of a Stated Preference (SP) study aimed at understanding the value that car drivers put on car sharing as opposed to single occupancy trips. Following an initial pilot period, 673 responses were received from a web-based survey conducted in June 2008 amongst a representative sample of car driving commuters in Scotland.

An important methodological aspect of this study was the need to account for differences in behaviour to identify those market segments with the greatest propensity to car share. To this end, we estimated a range of choice model forms and compared the ability of each to consistently identify individual behaviours. More specifically, this included a comparison of:

Standard market segmentation approaches based on multinomial logit with attribute coefficients estimated by reported characteristics (e.g. age, income, etc.);

A two-stage mixed logit approach involving the estimation of random parameters logit models followed by an examination of individual respondent's choices to arrive at estimates of their parameters, conditional on know distributions across the population (following Revelt & Train, 1999); and

A latent-class model involving the specification of C classes of respondent, each with their own coefficients, and assigning each individual a probability that they belongs to a given class based upon their observed choices, socioeconomic characteristics and their reported attitudes.

As hypothesised, there are significant variations in tastes and preferences across market segments, particularly for household car ownership, gender, age group, interest in car pooling, current journey time and sharing with a stranger (as opposed to family member/friend). Comparing the sensitivity of demand to a change from a single occupancy to a car-sharing trip, it can be seen that the latter imposes a ‘penalty’ equivalent to 29.85 IVT minutes using the mixed logit structure and 26.68 IVT minutes for the multinomial specification. Segmenting this latter value according to the number of cars owner per household results in ‘penalties’ equivalent to 46.51 and 26.42 IVT minutes for one and two plus car owning households respectively.

In this study, the authors examined demand-side credit in rural China with the aims of understanding attribute preferences and the willingness of farmers to pay for credit.

The authors implemented an in-the-field discrete choice experiment (DCE) using a D-optimal block (6 × 9 × 3) design applied to 420 farm households across five Chinese provinces (Shandong, Sichuan, Shaanxi, Jiangsu and Henan) in the summer and fall of 2018. The DCE included six attributes including the interest rate, term of loan, type of loan, type of repayment, type of institution and mobile banking services.

Conditional and mixed logit results indicated a downward sloping credit demand curve with variable elasticity across regions. Provincial willingness-to-pay (WTP) indicators suggested that farmers were willing to pay a premium for long-term ( 0.03–0.687%) and low collateral credit loans ( 0.79–2.93%). Also, four of five provinces indicated a preference for loan amortization rather than lump-sum payment. Interestingly, in comparison to the Agricultural Bank of China (ABC), only farmers in Shandong, Sichuan and Shaanxi indicated a preference for rural credit cooperatives (RCCs)/banks and the Postal Savings Bank of China (PSBC). Another quite surprising result was bank services, in our case, access to mobile banking did not appear to induce WTP for agricultural credit. While conditional and mixed logit regression coefficients were similar (and therefore robust), the authors found that there was substantial heterogeneity across attribute preferences on term of loan, type of loan and amortization. Preferences for type of lender and mobile banking were generally homogenous. This result alone suggested that lenders should consider offering a suite of credit products with different attributes in order to maximize the potential pool of borrowers. While there were some differences across provinces, farmers appeared to be indifferent to lenders, and it did not appear that offering banking services such as mobile banking had any bearing on credit decisions.

This paper presents a first step in using in-the-field choice experiments to better understand rural finance in China. Although the sample size satisfies conventional levels of significance and rank conditions, the authors caution against attributing results to China as a whole. Different provinces have different institutional structures and agricultural growing conditions and economies and these effects may differentially affect WTP for credit. Although by all indications farmers were aware of credit, not all farmers, in fact a minority, actually borrowed from a financial institution. This is not unusual in China, but for these farmers, the DCE was posed as hypothetical. Likewise, the study’s design was based on a generic credit product typical of rural China, and the authors caution against making inferences about other products with different attributes and risk structures.

This study is motivated by the rapidly changing dynamic in China's agricultural economy. With specific reference to new laws and regulations about the transfer of land use rights (LURs), China's agricultural economy is undergoing significant and rapid change which will require better understanding by policy makers, lenders and practitioners of the changing credit needs of farmers, including the new and emerging class of commercial farmers.

To the best of the authors’ knowledge, the authors believe that the result provided in this paper present the first use of in-the-field DCE and are the first to be reported in either the English or Chinese literature on rural credit product design.

Strategy deals with decisions about the scope of the firm and related choices about how to compete in various businesses. As such, research in strategy entails the analysis of discrete choices that may not be independent of each other. In this paper, we review the methodological implications of modeling such choices and propose conditional, nested, mixed logit, and hazard rate models as solutions to the issues that arise from non-independence among strategic choices. We describe applications with an emphasis on international strategy, an area where firms face a multiplicity of choices with respect to both location and mode of entry.

This paper aims to introduce and illustrate how discrete choice experiments (DCEs) can be used by accounting researchers and present an agenda of accounting-related research topics that might usefully benefit from the adoption of DCEs.

Each major phase involved in conducting a DCE is illustrated using a capital budgeting case study. The research agenda is based on a review of experimental research in financial accounting, management accounting and auditing.

DCEs can overcome some of the problems associated with asking decision-makers to rank or rate alternatives. Instead, they ask decision-makers to choose an alternative from a set. DCEs arguably better reflect the realities of real-world decision-making because decision-makers need to make trade-offs between all of the alternatives relevant to a decision. An important advantage that DCEs offer is their ability to calculate willingness-to-pay estimates, which can enable the valuation of non-market goods. Several streams of experimental accounting research would appear well-suited to investigation with DCEs.

While every effort has been made to ensure that this illustration is as generic to as the many potential studies as possible, it may be that researchers seeking to utilise a DCE need to refer to additional literary sources. This study, however, should serve as a useful starting point.

Accounting researchers are expected to benefit from reading this article by being: made aware of the DCE method and its advantages; shown how to conduct a DCE; and provided with an agenda of accounting-related research topics that might usefully benefit from application of the DCE methodology.

It is the authors’ understanding that this is the first article directed to accounting academics regarding the conduct of DCEs for accounting research. It is hoped that this study can provide a useful platform for accounting academics to launch further research adopting DCEs.

Mixed logit models can represent heterogeneity across individuals, in both observed and unobserved preferences, but require computationally expensive calculations to compute probabilities. A few methods for including error covariance heterogeneity in a closed form models have been proposed, and this paper adds to that collection, introducing a new form of a Network GEV model that sub-parameterizes the allocation values for the assignment of alternatives (and sub-nests) to nests. This change allows the incorporation of systematic (nonrandom) error covariance heterogeneity across individuals, while maintaining a closed form for the calculation of choice probabilities. Also explored is a latent class model of nested models, which can similarly express heterogeneity. The heterogeneous models are compared to a similar model with homogeneous covariance in a realistic scenario, and are shown to significantly outperform the homogeneous model, and the level of improvement is especially large in certain market segments. The results also suggest that the two heterogeneous models introduced herein may be functionally equivalent.

Patients and health professionals often make decisions which involve a choice between discrete alternatives. This chapter reviews the econometric methods which have been developed for modelling discrete choices and their application in the health economics literature. We start by reviewing the multinomial and mixed logit models and then consider issues such as scale heterogeneity, estimation in willingness to pay space and attribute non-attendance.

This article addresses simultaneously two important features in random utility maximisation (RUM) choice modelling: choice set generation and unobserved taste heterogeneity. It is proposed to develop and to compare definitions and properties of econometric specifications that are based on mixed logit (MXL) and latent class logit (LCL) RUM models in the additional presence of prior compensatory screening decision rules. The latter allow for continuous latent bounds that determine choice alternatives to be or not to be considered for decision making. It is also proposed to evaluate and to test each against the other ones in an application to home-to-work mode choice in the Paris region of France using 2002 data.

Research on strategic choices available to the firm are often modeled as a limited number of possible decision outcomes and leads to a discrete limited dependent variable. A limited dependent variable can also arise when values of a continuous dependent variable are partially or wholly unobserved. This chapter discusses the methodological issues associated with such phenomena and the appropriate statistical methods developed to allow for consistent and efficient estimation of models that involve a limited dependent variable. The chapter also provides a road map for selecting the appropriate statistical technique and it offers guidelines for consistent interpretation and reporting of the statistical results.

Purpose – This chapter gives an overview of methods for defining and analysing crash severity.

Methodology – Commonly used methods for defining crash severity are surveyed and reviewed. Factors commonly found to be associated with crash severity are discussed. Approaches for formulating and estimating models for predicting crash severity are presented and critiqued. Two examples of crash severity modelling exercises are presented and findings are discussed. Suggestions are offered for future research in crash severity modelling.

Findings – Crash severity is usually defined according to the outcomes for the persons involved. The definition of severity levels used by law enforcement or crash investigation professionals is less detailed and consistent than what is used by medical professionals. Defining crash severity by vehicle damage can be more consistent, as vehicle response to crash forces is more consistent than that of humans. Factors associated with crash severity fall into three categories – human, vehicle/equipment and environmental/road – and can apply before, during or after the crash event. Crash severity can be modelled using ordered, nominal or several different types of mixed models designed to overcome limitations of the ordered and nominal approaches. Two mixed modelling examples demonstrate better prediction accuracy than ordered or nominal modelling.

Research Implications – Linkage of crash, roadway and healthcare data sets could create a more accurate picture of crash severity. Emerging statistical analysis methods could address remaining limitations of the current best methods for crash severity modelling.

Practical Implications – Medical definitions of injury severity require observation by trained medical professionals and access to private medical records, limiting their use in routine crash data collection. Crash severity is more sensitive to human and vehicle factors than environmental or road factors. Unfortunately, human and vehicle factor data are generally not available for aggregate forecasting.