help fraclogit
-------------------------------------------------------------------------------

Title

fraclogit -- Fractional logit model as implemented in Wedderburn (1974) and generalized by McCullagh (1983).

Syntax fraclogit depvar indepvars [if] [in] [weight], [options]

options Description ------------------------------------------------------------------------- eform odds ratio form results Description

fraclogit carries out a quasi-likelihood estimation of a fractional logit model as described Wedderburn (1974) and generalized by McCullah (1983). The dependent variable for fraclogit is assumed to a proportion in the (0,1) interval. This model was rediscovered two decades later by Papke and Wooldridge (1996).

Example (from Wedderburn 1974)

. use wedderburn, clear . fraclogit yield i.site i.variety . margins variety

Estimation Details

fraclogit uses iteratively reweighted least squares in repeated calls to glm.

Saved Results

fraclogit saves the following in e():

Scalars e(N) number of observations e(df) number of model parameters e(dev) deviance e(rmse) root mse (dispersion)

Macros e(depvar) name of dependent variable Matrices e(b) 1 X K vector of estimates e(V) K X K variance-covariance matrix

Functions e(sample) marks estimation sample

Authors

Daniel A. Powers, University of Texas at Austin(dpowers@austin.utexas.edu).

References

McCullagh, P. (1983) "Quasi-Likelihood Functions," Annals of Statistics. Vol. 11, No. 1, pp. 59-67.

Papke, L.E. and J.W. Wooldridge (1996) "Econometric Methods for Fractional Response Variables With an Application to 401 (K) Plan Participation Rates," Journal of Applied Econometrics. Vol 11, No. 6, pp. 619-632.

Wedderburn, R.M.W. (1974) "Quasi-Likelihood Functions, Generalized Linear Models, and the Gauss-Newton Method," Biometrika, Vol. 61, No. 3, pp. 439-447.

Also see