Determination of optimal number of independent components in yield traits in rice

The principal component regression (PCR) and the independent (ICR) are dimensionality reduction methods extremely important in genomic prediction. These require choice of number components to be inserted into model. For PCR, there formal criteria; however, for ICR, adopted criterion chooses (ICs) associated greater accuracy requires high computational time. In this study, seven criteria based on (PCs) variable selection guide ICR proposed evaluated simulated real data. both datasets, most efficient that drastically reduced time determined ICs should equal PCs reach a higher value. addition, did not recover heritability generated biased values.