New statistical test for quality control in high dimension data set

• Main Authors: Sharif, Shamshuritawati, Ismail, Suzilah, Omar, Zurni
• Format: Article
• Language: English
• Published: Research India Publications 2017
• Subjects: QA75 Electronic computers. Computer science
• Online Access: http://repo.uum.edu.my/27955/1/IJAER%2012%2016%202017%206241%206249.pdf; http://repo.uum.edu.my/27955/; http://www.ripublication.com/Volume/ijaerv12n16.htm

The existing statistical test for quality control such as Box’s M test caters for number of sample size n larger than number of variables or dimensions p (n>p). However, in real world application of Small Medium Enterprise (SME), the number of variables or dimensions p can be larger than sample size n (p>n) which is known as high dimension data set. This is due to small number of daily productions which lead to small number of sample size (n) but high dimensions products (p). One of the examples is rubber gloves which rely on machine capacity or latex supply that limiting the daily productions (n) but involves many dimensions (p) of measurement such as the size of five different fingers, the width of the palm and wrist; the strength, number of holes and etc. Another drawback, once the samples are tested for quality control, they are discarded which is very wasteful if uses large sample size. Therefore, in this study we have developed a new statistical test known as S* test to accommodate high dimension data set (p>n). A simulation study was conducted in comparing the performance of Box’s M test and S* test using power of test. Based on 10,000 replications and 5% significance level, the power of test indicated that S* test outperformed the Box’s M test. Interestingly, when n is smaller than p, S* test still can be computed which proven it can be used for quality testing in high dimension data set.