The purpose of this report is to review the potential for applying some recent advances in econometrics, particularly in the analysis of categorical data, to road accident research. This report should interest: (a) those involved in the analysis of road accident data; (b) those involved in road accident research; (c) statisticians and econometricians with an interest in the analysis of categorical data. The major conclusions are that the set of statistical techniques termed quantal response models, and extensions to these techniques, appear especially suited to the analysis of road accident data in that they possess an intuitively appealing theoretical framework, they can be estimated using individual data, thus providing certain statistical advantages and survey economies, and they offer considerable flexibility. (A)
Abstract