By all, we are done with the computation of PCA in R. Please be aware that biopsy_pca$sdev^2 corresponds to the eigenvalues of the principal components.įinally, the last row, Cumulative Proportion, calculates the cumulative sum of the second row. For other alternatives, see missing data imputation techniques.īiopsy_pca$sdev ^ 2 / sum (biopsy_pca$sdev ^ 2 ) # 0.655499928 0.086216321 0.059916916 0.051069717 0.042252870 # 0.033541828 0.032711413 0.028970651 0.009820358Īccordingly, the first principal component explains around 65% of the total variance, the second principal component explains about 9% of the variance, and this goes further down with each component. We will also exclude the observations with missing values using the na.omit() function to keep it simple. We will exclude the non-numerical variables before conducting the PCA, as PCA is mainly compatible with numerical data with some exceptions. The output also shows that there’s a character variable: ID, and a factor variable: class, with two levels: benign and malignant. Īs shown below, the biopsy data contains 699 observations of 11 variables. # $ class: Factor w/ 2 levels 'benign', # 'malignant': 1 1 1 1 1 2 1 1 1 1.
