REG2LOGIT: Stata module to approximate logistic regression parameters using OLS linear regression

reg2logit estimates the parameters of a logistic regression of yvar on xvars by transforming OLS estimates of the linear regression of yvar on xvars. Factor xvars are allowed. The transformation formula, first derived by Haggstrom (J.Bus.Econ.Stat., 1983), is discussed by Allison (2020). The transformed OLS estimates are fully efficient estimates of the logistic regression under the assumption that the xvars are multivariate normal conditionally on the value of the yvar. If the xvars are in fact conditionally multivariate normal, then the estimates produced by reg2logit are more efficient than the "distribution-free" estimates produced by the logit command, which assume nothing about the distribution of the xvars.