Optimal model averaging estimator for multinomial logit models

In this paper, we study optimal model averaging estimators of regression coefficients in a multinomial logit model, which is commonly used many scientific fields. A Kullback–Leibler (KL) loss-based weight choice criterion developed to determine weights. Under some regularity conditions, prove that the resulting are asymptotically optimal. When true one candidate models, averaged consistent. Simulation studies suggest superiority proposed method over selection criterions, methods, as well other related methods terms KL loss and mean squared forecast error. Finally, website phishing data illustrate method.