What Can We Learn about Correlations from Multinomial Probit Estimates

摘要: It is well known that, in a multinomial probit, only the covariance matrix of the location and scale normalized utilities are identified. In this study, we explore the relation between these identifiable parameters original elements of the covariance matrix, to find out what can be learnt about correlations between the stochastic components non-normalized utilities. We show certain circumstances, it possible obtain information on behavioural parameters and define appropriate tools for inference. illustrate usefulness our results in applied settings using an example.