Symmetricity between the sampling distribution of coefficient of variations, CVc and CVr for Gumbel samples.

• Main Authors: Maarof, Fauziah, Mohamed, Shamsiah, Ahmad, Aimi Athirah
• Format: Conference or Workshop Item
• Language: English; English
• Online Access: http://psasir.upm.edu.my/id/eprint/27619/1/ID%2027619.pdf; http://psasir.upm.edu.my/id/eprint/27619/

A robust form of the coefficient of variation (cv). CVr • constructed based on the sample median absolute deviation. MAD and the sample median is considered. Histograms. boxplolS and descriptive statistics of the sampling distributions of CVr are generated for sample from Gumbel distribution with parameter value (u = I, a = 0.05) and sample sizes (n =20. 50. 100, 300. 500). The histograms, boxplolS, measure of skewness and kunosis observed indicated thallhe sampling distribution of CVr becomes more symmetric and nonnal as the sample size increases. outperforming the sampling distribution of the conventional coefficient of variation, eVe with respect to the degree and rate towards symmetricity. It is also observed that the sampling distribution of the CVr tends towards normality faster and it has smaller kultosis compared to that of the sampling distribution of CVc.