Hi, I transformed the variables in Z scores, and then, when I analyze with principal component analysis (PCA), we have table of rotated component matrix (with 2 parts: Raw and Rescaled), even in variance table, there is these two parts. So, which one must be considered
Dear Francois,
rescaled. Ok rescaled meant z-scores
Hi Francois,
I need to know more about the reasons for doing a pricipal components analysis for your project. Why did you have to transform the raw data to z scores? As Beatrice' said above, most likely you need to use the rescaled results since you must have had good reasons for transforming the data.
Good luck.
Raid
If you are doing numerical transformations to z rather than area transformations, both results should be the same if the appropriate metric is used to interpret each. Thus, I do not think it makes a difference whether you use raw or rescaled if you interpret appropriately. You should get the same principal components and account for the same proportion of variance.
Dear Francois
You can give more details about your study including the aim of the study and your hypothesis , the distribution of your data.
Good Luck
the reason to transform the variables in Z scores, are that, the values of each parameter is different to onather one. we analyze the heavy metals and physicochemical parameters of soils. So the values are differents.
and about the Raw and Rescaled, there are equal some how, there is not a big differency! So asked the question in order to exactly which one is using.
Dear Francois
You can use transformation to a percentage of the mean instead of Z score , for example, to compare variance in two groups different in the units we use the coefficient of variation ( C. V. % ), this approach cancel the role of the units and use the percentage in the calculation.
Good Luck
- Badges
- Science topic
- Similar topics
- Medicine
- Physiology
- Biosignals
- Biomedical Signal Processing
- Statistical Analysis