Logistic回归模型下的一些岭参数性质

his paper introduces some new different methods for estimating the ridge parameter k for logistic ridge regression (LRR) model by using the maximum likelihood (ML) method. The performance of these new approaches are evaluated and compared through Monte Carlo simulations study along with some other ridge of k in terms of mean square error (MSE), the mean value and the standard deviation, respectively. The results from the simulation study show that LRR outperforms ML approach, and some new proposed ridge estimators outperform those proposed by Kibria et al. 2012which are to be very reasonable alternatives in practice.