Betit: A Family that Nests Probit and Logit
Published By: IZA on eSS | Published Date: December, 13 , 2000This paper proposes a dichotomous choice model that is based on a transformed beta (or z”)
distribution. This model, called betit, nests both logit and probit and allows for various skewed
and peaked disturbance densities. Because the shape of this density affects the estimated
relation between the dichotomous choice variable and its determinants, the greater flexibility of
the transformed beta distribution is useful in generating more accurate representations of this
relationship. The paper considers asymptotic biases of the logit and probit models under
conditions where betit should have been used. It also investigates small sample power and
provides two examples of applications that illustrative of the capability of the betit model. [IZA Discussion Paper No. 222]
Author(s): Wim P.M. Vijverberg | Posted on: Aug 13, 2010 | Views(1144) | Download (1219)