A0243
Title: Discriminant analysis based on binary time series
Authors: Yuichi Goto - Kyushu University (Japan) [presenting]
Masanobu Taniguchi - Waseda University (Japan)
Abstract: Binary time series is the time series taking values 0 and 1. We discuss discriminant analysis and propose a new classification method based on binary time series. First, we show that the misclassification probability tends to zero when the number of observations tends to infinity. Next, we evaluate the asymptotic misclassification probability when two categories are contiguous. Finally, we show that our classification method based on binary time series has good robustness when the process is contaminated by an outlier, that is, our classification method is insensitive to the outlier. However, the classical method based on smoothed periodogram is sensitive to the outlier.
