A New Approximation for the Null Distribution of the Likelihood Ratio Test Statistics for k Upper Outliers in a Normal Sample

Mahmoudvand, Rahim and Hassani, Hossein (2005): A New Approximation for the Null Distribution of the Likelihood Ratio Test Statistics for k Upper Outliers in a Normal Sample. Published in: Journal of Statistical Research of Iran No. 2 (1 September 2006): pp. 155-161.

Abstract

In this paper, we introduce a new approximation for the null distribution of the likelihood ratio test for the general case. We compare the the critical values obtained by the new approximation to the values which are obtained by the exact distribution for the cases k=1, 2 to test the accuracy of the new approximation. Also, we compare the results to another approximation method (which is known by Barnett and Lewis (1994)) for the cases k=3,4.

• Item Type: MPRA Paper
• Language: English
• Keywords: Likelihood Ratio Test, Outlier, Approximation, Normal Sample
• Subjects: C - Mathematical and Quantitative Methods > C0 - General > C02 - Mathematical Methods; Z - Other Special Topics > Z0 - General > Z00 - General

References

Barnett, V.; Lewis, T. (1994). Outliers in Statistical Data. 3rd ed. John Wiley & Sons, Chichester.

Thode, Jr.(2002). Testing for Normality. Marcel Dakker Inc. New Yourk.

Zhang, J.; Yu, K. (2006). The Null Distribution of the Likelihood-Ratio Test for One or Two Outliers in a Normal Sample. Test,Vol. 15. No 1. pp. 141--150.