New approach application of data transformation in mean centering of ratio spectra method

• Mean centering of ratio spectra (MCR) based on arithmetic mean.
• New approach of data transformation in MCR spectra based on geometric mean.
• Geometric mean is a better measure of central tendency than arithmetic mean.
• Logarithmic transformation and geometric mean were subjected to skewed data.
• Applied to resolve ASP, ATOR and CLOP in synthetic mixture and pharmaceuticals.

Most of mean centering (MCR) methods are designed to be used with data sets whose values have a normal or nearly normal distribution. The errors associated with the values are also assumed to be independent and random. If the data are skewed, the results obtained may be doubtful. Most of the time, it was assumed a normal distribution and if a confidence interval includes a negative value, it was cut off at zero. However, it is possible to transform the data so that at least an approximately normal distribution is attained. Taking the logarithm of each data point is one transformation frequently used. As a result, the geometric mean is deliberated a better measure of central tendency than the arithmetic mean. The developed MCR method using the geometric mean has been successfully applied to the analysis of a ternary mixture of aspirin (ASP), atorvastatin (ATOR) and clopidogrel (CLOP) as a model. The results obtained were statistically compared with reported HPLC method.

Figure optionsDownload as PowerPoint slide