Data Transformation Technique to Improve the Outlier Detection Power of Grubbs' Test for Data Expected to Follow Linear Relation

Grubbs test (extreme studentized deviate test, maximum normed residual test) is used in various fields to identify outliers in a data set, which are ranked in the order of . However, ranking of data eliminates the actual sequence of a data series, which is an important factor for determining outliers in some cases (e.g., time series). Thus in such a data set, Grubbs test will not identify outliers correctly. This paper introduces a technique for transforming data from sequence bound linear form to sequence unbound form . Applying Grubbs test to the new transformed data set detects outliers more accurately. In addition, the new technique improves the outlier detection capability of Grubbs test. Results show that, Grubbs test was capable of identifing outliers at significance level 0.01 after transformation, while it was unable to identify those prior to transforming at significance level 0.05.

[1]  W. Fung.  A statistical-test-complemented graphical method for detecting multiple outliers in two-way tables , 1991 .

[2]  F. E. Grubbs.  Sample Criteria for Testing Outlying Observations , 1950 .

[3]  Shaochun Wu,et al.  Conflict Analysis of Multi-source SST Distribution , 2009, HPCA.

[4]  Marek K. Solak,et al.  DETECTION OF MULTIPLE OUTLIERS IN UNIVARIATE DATA SETS , 2009 .

[5]  M. Srivastava.  Effect of equicorrelation in detecting a spurious observation , 1980 .

[6]  F. E. Grubbs.  Procedures for Detecting Outlying Observations in Samples , 1969 .

[7]  Virgil R. Marco,et al.  On the effect of correlation and unequal variances in detecting a spurious observation , 1989 .

[8]  Bernard Rosner,et al.  On the Detection of Many Outliers , 1975 .

[9]  B. Rosner.  Percentage Points for a Generalized ESD Many-Outlier Procedure , 1983 .

[10]  Ram B Jain,et al.  A recursive version of Grubbs' test for detecting multiple outliers in environmental and chemical data. , 2010, Clinical biochemistry.

[11]  S. Puntanen,et al.  A complete solution to the problem of robustness of Grubbs's test , 1990 .

[12]  Michael Thompson,et al.  Notes On Statistics And Data Quality For Analytical Chemists , 2011 .

[13]  O. Christie,et al.  Data Transformation as a Means to obtain Reliable Consensus Values for Reference Materials , 1977 .

[14]  F. E. Grubbs,et al.  Extension of Sample Sizes and Percentage Points for Significance Tests of Outlying Observations , 1972 .

[15]  R. Brant,et al.  Comparing Classical and Resistant Outlier Rules , 1990 .

[16]  Seymour Geisser,et al.  Influential observations, diagnostics and discovery tests , 1987 .

[17]  Rong Pan,et al.  A Sequential Markov Chain Monte Carlo Approach to Set-up Adjustment of a Process over a Set of Lots , 2004 .