3. See Ben Dor, Dynkin, Houweling, Hyman, van Leeuwen, and Penninga, “DTS (Duration Times Spread): A New Measure of Spread Exposure in Credit Portfolios” for further detail regarding the analyses shown in Exhibits 55–2 and 55–3. In this section, we report our original results using data through 2005. In what follows, we use market data after this date as an extreme out-of-sample test of these relationships.

4. Bonds were assigned to one of 66 cells with breakpoints ranging from 100 bp to more than 500 bp (the financial, industrial and utility sectors had seven, nine, and six spread cells for each maturity bucket, respectively). To eliminate idiosyncratic effects, months in which a cell was populated by fewer than 20 bonds were excluded. Employing a weighted volatility estimate instead (in which the number of observations in each month was used as the weighing factor) produced very similar results.

..................Content has been hidden....................

You can't read the all page of ebook, please click here login for view all page.
Reset
44.215.110.142