I want ti#o know the best method to use in determining the number of component to retain in Principal component analysis. Between the Kaiser rule, Scree plot and parallel analysis, which is the most accurate to use please.
There is not a single method that is best. That's why there are so many approaches to this problem. My suggestion would be to use them all, study the results, what do they have in common, where are the differences. Last but not least, you may also want to take the interpretability of principal components into account.