<table cellspacing="0" cellpadding="0" border="0" ><tr><td valign="top" style="font: inherit;"><div id="yiv185778574">Hi!<br><br>Here is one small clarification I am seeking regarding the Portfolio probability distribution (where no of Obligors are 2). It is the classical example as given in the CreditMetrics document.<br><br>BOND "BBB"<br><br>Possible Migrations AAA AA A BBB BB B CCC Default<br><br>Probabilities(%) 0.02 0.33 5.95 86.93 5.30 1.17 0.12 0.18 <br><br>Year End
values* 109.37 109.19 108.66 107.55 105.02 98.10 83.64 51.13<br><br><br>Mean(BBB) = 107.09 (= mu1)<br>sd(BBB) = 2.99 (= sd1)<br><br>* These are the values taken by BBB bond after migrating to respective ratings after 1 year.<br><br>Similarly for Bond "A"<br><br>Possible Migrations AAA AA A BBB BB B CCC Default<br><br>
Probabilities(%) 0.09 2.27 91.05 5.52 0.74 0.26 0.01 0.06 <br>
Year End values* 106.59 106.49 106.30 105.64 103.15 101.39 78.71 51.13<br>
<br><br>Mean(A) = 106.55 (= mu2)<br>sd(A) = 1.49 (= sd2) <br><br><br><br>#### Case 1 :- The correlation between BBB and A is 0.<br><br>So Probability that after 1 year, BBB will remain BBB and A will remain A is 0.8693*0.9105 = 0.7915<br><br><br><br>#### Case 2 : Let the correlation between BBB and A is 0.30 (= r )<br><br>So Probability that after 1 year, BBB will remain BBB and A will remain A = 0.7969<br>
<br>I have tried to arrive at this value but somehow I am missing something. I have used the following formula (assuming the normalized returns on two assets BBB and A)<br><br><br>p(BBB, A) = 1/[2*pi*sd1*sd2*sqrt(1-r^2)] * [exp (-z / 2*(1 - r^2))] where <br><br>z = ((x1-mu1)/sd1)^2 + ((x2 - mu2)/sd2)^2 - 2 * r * (x1-m1)/sd1 * (x2-mu2) / sd2<br><br>(formula available at http://mathworld.wolfram.com/BivariateNormalDistribution.html)<br><br><br>I have taken x1 = 107.55 and x2 = 106.30 and the probability value I arrive at using above formula = 0.03612<br><br>The actual answer is 79.69% i.e. 0.7969 as given in the Creditmetrics technical document (page 26 - table 1.7).<br><br><br>I am held up at this particular point. I will be grateful if someone can guides me. Also I am not sure I will be able to attach the excel file with this mail, but nevertheless I am trying the same. I sincerely apologize for writing such a long
mail.<br><br><br>Regards<br><br>Milano<br><br><br><span style="font-size: 9pt; line-height: 150%; font-family: Verdana;" lang="EN-US"><p>
</p><p align="center"><br></p></span><div style="text-align: left;"><span style="font-size: 9pt; line-height: 150%; font-family: Verdana;" lang="EN-US"><p align="center"><br></p></span><br><span style="font-size: 9pt; line-height: 150%; font-family: Verdana;" lang="EN-US"></span></div><span style="font-size: 9pt; line-height: 150%; font-family: Verdana;" lang="EN-US"><p align="center"><img alt=""></p></span><span style="font-size: 9pt; line-height: 150%; font-family: Verdana;" lang="EN-US"><p align="center"><br></p><p align="center"><br></p><p align="center"><br></p></span><br></div></td></tr></table><br>