Posters
« Back
Introducing Undergraduate Chemists to Chemometrics, Part 2: Performing Outlier Rejection Tests in a Spreadsheet Environment
EP25656
Poster Title: Introducing Undergraduate Chemists to Chemometrics, Part 2: Performing Outlier Rejection Tests in a Spreadsheet Environment
Submitted on 31 Mar 2017
Author(s): Mark T. Stauffer
Affiliations: University of Pittsburgh - Greensburg
This poster was presented at Pittcon 2017
Poster Views: 948
View poster »


Poster Information
Abstract: This presentation is a continuation of work initiated by the author to introduce undergraduates majoring in chemistry, and also biology and other technical fields, to the field of chemometrics. With the widespread availability of instruments capable of collecting enormous amounts of data, and the software capable of analyzing that data toward analyte quantitation, it is highly important for undergraduate chemists to know and understand the principles and operations of various chemometric methods. Additionally, undergraduate chemists need to learn and understand various statistical tools (which are part of the field of chemometrics, too) for such data analysis tasks as rejection of suspect outlier data and results, among others. Software packages, e.g., Microsoft® Excel™ and MATLAB®, facilitate the performance of the plethora of mathematical and statistical methods for data and results analysis – and undergraduate chemists need to learn these tools as well.

In this paper, the author will present some handy techniques for setting up and implementing Microsoft® Excel™ worksheets for performance of outlier rejection tests, e.g., the Dixon Q, Grubbs G, and modified Thompson tau tests. An additional outlier test for x,y data pairs will be presented to round out the presentations of outlier tests using Excel™. Background on the outlier rejection tests will be presented, and the advantages and disadvantages of each test will be discussed. Specific examples utilizing synthetic and real data will be presented to illustrate how Excel™ may be used for each outlier rejection test. Attempts to streamline the NIPALS (nonlinear iterative partial least squares) algorithms for principal component analysis (PCA) and partial least squares regression (PLSR) presented at the last Pittsburgh Conference, were met with some obstacles during the past year. The author is continuing work toward streamlining the NIPALS algorithm in Excel™, with expectations of presentation of that research at Pittcon 2018.
Summary: This presentation presents and discusses how the author is using a well known spreadsheet program to introduce undergraduate chemistry majors, and students majoring in other technical disciplines, to outlier data rejection/retention in chemical analysis.References: [1] Dean, R. B. and Dixon, W. J. "Simplified Statistics for Small Numbers of Observations". Anal. Chem. 1951, 23(4), 636–638.
[2] Grubbs, F. E. (1950). "Sample criteria for testing outlying observations". Annals of Mathematical Statistics 1950, 21(1), 27–58.
[3] Harris, D. C. Quantitative Chemical Analysis, 9th edition; 2003, New York: W.H. Freeman and Co.
[4] Thompson .R. "A Note on Restricted Maximum Likelihood Estimation with an Alternative Outlier Model". Journal of the Royal Statistical Society, Series B (Methodological) 1985, 47(1), 53-55.
[5] “Outliers”, J. M. Cimbala, Penn State Univ., latest revision Sept. 12, 2011; 5 pp (http://www.mne.psu.edu/cimbala/me345/Lectures/Outliers.pdf, accessed December 17, 2014).
[6] Anbarasi, M.S. et al. “Outlier Detection for Multidimensional Medical Data”. International Journal of Computer Science and Information Technologies (IJCSIT), 2011, 2(1), 512-516.
Report abuse »
Questions
Ask the author a question about this poster.
Ask a Question »

Creative Commons

Related Posters


Identification using Forensic Odontology
Vyom Rathi

Accelerated Ageing in Depression: A Study of Two Cohorts
Mathew A. Harris, Laura de Nooij, Xueyi Shen, Toni-Kim Clarke, Riccardo Marioni, Simon R. Cox, Emma L. Hawkins, Mark J. Adams, Liana Romaniuk, Stephen M. Lawrie, James H. Cole, Andrew M. McIntosh and Heather C. Whalley

Digiceuticals
Helana Lutfi and Shaban Nuredini

Machine Learning: A New Breakthrough in Medical Diagnosis
Punitha Mahendran, Anis Joelaira, Chai Yong Chia, Fazidatul Aziz, Hasyimah Emran, Siew Fun Lee, Wai Kit Wong, Yee Yan Tang, Li Yang Wong

Accreditation For a Better Medical Laboratory Quality: ISO 15189:2012
Lee Siew Fun, Tay Yi Wen, Muhammad Hafizan Mustari, Wong Sweet Mun, Nur'Awatif Ishak