This is a preview. Log in through your library . Abstract This paper rigorously establishes that the existence of the maximum likelihood estimate (MLE) in high-dimensional logistic regression models ...

We consider estimation of mixed-effects logistic regression models for longitudinal data when missing outcomes are not missing at random. A typology of missingness mechanisms is presented that ...

In recent columns we showed how linear regression can be used to predict a continuous dependent variable given other independent variables 1,2. When the dependent variable is categorical, a common ...

Modern statistical modelling is increasingly focused on reducing bias and enhancing the accuracy of parameter estimates. Traditional maximum likelihood estimation, while powerful, can encounter ...