Yes, it is necessary to normalize data before performing PCA. ... If you normalize your data, all variables have the same standard deviation, thus all variables have the same weight and your PCA calculates relevant axis.
Why does PCA need standardization?
The reason why standardization is very much needed before performing PCA is that PCA is very sensitive to variances. Meaning, if there are large differences between the scales (ranges) of the features, then those with larger scales will dominate over those with the small scales.
How do you standardize data for PCA?
There are 6ish steps to PCA:
- Standardize data.
- Construct covariance matrix.
- Extract eigenvectors and eigenvalues from the covariance matrix.
- Sort the eigenvalues (and their eigenvectors!) in decreasing order.
- Select a number of components to care about (and keep)
- Transform your dataset.
