Financial correlation matrices measure the unsystematic correlations betweenstocks. Such information is important for risk management. The correlationmatrices are known to be ``noise dressed''. We develop a new and alternativemethod to estimate this noise. To this end, we simulate certain time series andrandom matrices which can model financial correlations. With our approach,different correlation structures buried under this noise can be detected.Moreover, we introduce a measure for the relation between noise andcorrelations. Our method is based on a power mapping which efficientlysuppresses the noise. Neither further data processing nor additional input isneeded.
展开▼
