ASUMAN TURKMEN

My research interests are multivariate statistical methods that deal with robust estimation and outlier detection which has important applications in the fields of fraud detection, astronomy, bioinformatics (e.g., microarray experiments), and many other countless areas. Most statistical methods assume homogeneous (outlier-free) data in which all data points satisfy the same model. However, real data are not homogeneous; and accurate identification of outliers plays an important role in statistical analysis. My studies have been heavily concentrated on both outlier identification and robust estimation in regression, classification, PCA (principal component analysis), PLS (partial least squares), and ICA (independent component analysis). I was recently a part of an NSF project in which we proposed a new nonparametric classifier using the "data depth" concept that can be used to measure the "depth" or "outlyingness" of a given multivariate sample with respect to its underlying distribution.

Currently, I am working on robust partial least squares regression and classification methods for high-dimensional data, with applications in chemometrics and bioinformatics. Another topic that I am currently interested in is the variable selection via partial least squares method.