Beginner here. I’m trying to calculate a state’s infrastrucure index using different variables and I applied PCA.

At first I did a dot product of original data and the principal components.

pca = PCA(2).fit(ind_scaled)
ind_pca = pd.DataFrame(pca.components_)

pca_data = pd.DataFrame((np.dot(ind_scaled, pca.components_.T)).sum(axis=1))

index = pd.DataFrame()
index["States"] = states
index["Index"] = pca_data

Later I tried to use fit_transform

pca = PCA(2).fit(ind_scaled)
ind_pca = pca.transform(ind_scaled)

index = pd.DataFrame()
index["States"] = states
index["Index"] = ind_pca.sum(axis=1)

I got different answers for both. I’m not exactly sure how either results are different. Any kind of help is appreciated. Thank you