On the Information Content of Decomposed Financial Return Series: A Wavelet Approach

We decompose financial return series via wavelets into different time scales to analyse their information content regarding the volatility of the returns. Moreover, we investigate the information of each scale and discuss the decomposition of daily Value-at-Risk (VaR) forecasts. By an extensive empirical analysis, we analyse financial assets in calm and turmoil market times and show that daily VaR forecasts are mainly driven by the volatility which is captured by the scales comprising the short-run information. Further, we apply Extreme-Value-Theory to each time scale and illustrate that the information which is stored by the short-run scales linked via copulas, outperforms classical parametric VaR approaches which incorporate all information available.