Data Science Seminar
Date and Time: Wednesday, 30 November 2022, 3:30 pm
Venue: Lecture Hall 4
Detection of Multiple Outliers in Multidimensional Data
M.R. Srinivasan
Formerly Professor of Statistics, University of Madras.
30-11-22

---

Abstract

Statisticians have been interested in finding "outlying" or "unusual" observations for many years. Data that have been incorrectly entered or that do not belong to the population from which the rest of the data came can bias estimates and give misleading results. Traditional and non-traditional methods have been devised to identify and/or accommodate outlying observations in a variety of situations. With recent advances in technology, scientists are collecting larger data sets, and the analyst is getting deeper to unravel the mysteries of data

There are two basic reasons to search for outliers: (i). because there is interest in the outliers for their own sake, for example discovery of Argon and (ii). because the outliers could influence the results from the rest of the data. The presentation is an attempt to address the following objectives: Reviewing the existing methods and suggest an alternative method called Co-median method with better properties.