Reconstructing equation of state of dark energy with principal component analysis

Abstract: We represent a nonparametric method to reconstruct the equation of state for dark energy directly from observational Hubble parameter data. We use principal component analysis (PCA) to extract the signal from data with noise. Moreover, we modify Akaike information criterion (AIC) to guarantee the quality of reconstruction and avoid over-fitting simultaneously. The results show that our method is robust in reconstruction of dark energy equation of state. Although current observational Hubble parameter data alone can not give a strong constraint yet, our results indicate that future observations can significantly improve the quality of the reconstruction.