Reconstruction of Late-Time Cosmology Using Principal Component Analysis

We reconstruct late-time cosmology in a model-independent manner using the Principal Component Analysis (PCA) technique. We propose a variant of PCA that can be used to find out the functional form of late-time cosmological quantities. A combination of the PCA algorithm and the calculation of correlation coefficients are used as prime reconstruction tools. We carry out the analysis with simulated data as well as with real data. The correlation coefficient calculation also enables us to determine the final number of principal components we must keep in the final reconstruction process. The reconstruction indicates a slowly varying equation of state of dark energy.