Title: Asymptotic distribution of the linear discriminant function with two-step monotone sample
Authors: Nobumichi Shutoh - Kobe University (Japan) [presenting]
Abstract: The aim is to derive the asymptotic distribution of the linear discriminant function constructed by the estimators on the basis of monotone sample. Because the estimators of the parameters have more complicated forms as the number of missing patterns $k$ increases, we show the results for the case of $k=2$ for simplicity. The main results can be applied to approximate the misclassification probabilities for the linear discriminant function. Monte Carlo simulation is conducted in order to evaluate the accuracy of the approximation.