This code estimates both static and dynamic Logit and Probit (binary-choice) models. Three types of dynamic models can be considered by including the lagged binary variable (Dynamic_y), the lagged index (Dynamic_π) or both of them (Dynamic_y ; Dy_π) as explanatory variables. A (T,1) binary vector must be specified, as well as a (T,N) matrix of explanatory variables. The user can also choose the lag length, the type of model and select between Logit and Probit specifications. The results include estimated parameters, standard-errors, t-statistics as well as information criteria. The series of estimated probabilities is saved in .csv format.
Working Paper (2021)
Couperier Ophélie and Leymarie Jérémy