清華大學、交通大學
統 計 學 研 究 所
專 題 演 講
題 目: Imputation methods for microarray data
主講人: 袁新盛博士 (中央研究院 統計所)
時 間: 96年1月5日（星期五）10：40 - 11：30
（上午10：20-10：40茶會於統計所821室舉行）
地 點: 清大綜合三館837室
Abstract
Missing valuesa are often the obstacles for carrying out the statistical
analysis. Several commonly used methods, such as Singular Value Decomposition
and Principal Component Analysis, require the complete data set. To accommodate
this, researchers often preprocess the data through statistical imputation
methods. In microarray data analysis, Row Average and KNN are two widely used
methods. In this talk, we will introduce several novel imputation methods.
The core difference between the previous methods and our methods is that we
take the dependency between slides into consideration. The supporting
argument for this consideration is that the prediction of missing values
could be naturally derived from the repeat experiment. The methods and
comparisons will be covered in this talk.
--
※ 發信站: 批踢踢實業坊(ptt.cc)
◆ From: 140.113.114.192