3 papers
stat.ME2026
Maximum softly penalised likelihood in factor analysis
Philipp Sterzinger, Ioannis Kosmids, Irini Moustaki
Estimation in exploratory factor analysis often yields estimates on the boundary of the parameter space. Such occurrences, known as Heywood cases, are characterised by non-positive…
math.ST2026
Diaconis-Ylvisaker prior penalized likelihood for logistic regression
Philipp Sterzinger, Ioannis Kosmidis
We characterise the behavior of the maximum Diaconis--Ylvisaker prior penalized likelihood estimator in high-dimensional logistic regression, where the number of covariates is a fr…
stat.ME2024
Jeffreys-prior penalty for high-dimensional logistic regression: A conjecture about aggregate bias
Ioannis Kosmidis, Patrick Zietkiewicz
Firth (1993, Biometrika) shows that the maximum Jeffreys' prior penalized likelihood estimator in logistic regression has asymptotic bias decreasing with the square of the number o…