Date:         Tue, 21 Jul 2009 07:58:43 -0500
Reply-To:     SAS Kid <sasloving.kid@GMAIL.COM>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         SAS Kid <sasloving.kid@GMAIL.COM>
Subject:      Re: Credit risk - probability of default question
Comments: To: Philip Rack <PhilRack@minequest.com>
In-Reply-To:  <003201ca09f9$2cabbe40$86033ac0$@com>
Content-Type: text/plain; charset=ISO-8859-1

thank you Philip
is there an example anywhere
I wonder why we need to use the transition matrix then


On 7/21/09, Philip Rack <PhilRack@minequest.com> wrote:
> I don't believe you can get to the loan level this way. I don't know the
> loan types that you are analyzing but for sake of simplicity, look at
> mid-sized business loans. The PD is calculated based on attributes such as
> the number of days past 30 that the business has past-due accounts
> receivables, liquid assets available, owners and cosigners FICO Scores,
> etc...
>
> Those attributes are scored and added together (i.e. past-due AR's might
> account for 30% of the total PD) and then used in a model. Once you have an
> individual businesses PD, then you can look at the entire portfolio (or
> segment) to determine businesses at-risk and LGD.
>
> Phil
>
> Philip Rack
> MineQuest, LLC
> SAS & WPS Consulting and WPS Reseller
> Tel: (614) 457-3714
> Web: www.MineQuest.com
> Blog: www.MineQuest.com/WordPress
>
>
> -----Original Message-----
> From: SAS(r) Discussion [mailto:SAS-L@LISTSERV.UGA.EDU] On Behalf Of SAS Kid
> Sent: Monday, July 20, 2009 10:34 PM
> To: SAS-L@LISTSERV.UGA.EDU
> Subject: Credit risk - probability of default question
>
> I have a Probability of default question
>
> I am currently using a delinquency transition matrix to see how
> delinquencies migrate from
> 0 to 30 days, 30days to 60days, 60 to 90, 90 to 120 etc. I have a 5 by 5
> matrix and i have row percentages
> in each cell to see how probability of default migrates month to month.
> I use proc freq to get this matrix.
>
> My row percent in each cell
> ( which i call probability of default) = count in each cell/total number of
> loans in that row.
>
> The probabilities which i get from this matrix is at the segment level.
> However i like to compute probability of default at the loan level. I
> wonder how
> i can get probability of default at the loan level ? ultimately this is
> going to help me with expected loss calculations. I understand different
> products could have different default definitions.
>
> Any reference or help is appreciated.
>
> Thank you.
>
>