Identifying Risk Sources with Principal Component Analysis

Risk is an important characteristic that plays a huge role in capital markets decision making. Thus, correctly determining and specifying the risk sources on a given market may shape investors' decisions. This thesis aims to identify and interpret the sources of financial risk on the German equity market. The goal is achieved using principal component analysis as a tool for building the portfolios that allocate their budgets to distinct uncorrelated risk sources. Throughout the research, correlation matrix was compared with the covariance matrix as an underlying foundation for PCA, effectively examining if standardising the returns is an important step. It was also revealed that the first five principal components have different interpretations, indicating the underlying equity characteristics that may not be obvious on a first sight, effectively revealing the biggest risk sources on the German equity market. The market component was extracted that accounted for over 30% of the overall variance, representing a big portion of systematic risk. The findings of the study also suggest that normalising the returns might be a beneficial approach.