Harmoney

[RMJH]

Thoughts on this style:

Attachment 10063

That one I really like. Big picture and time based. And as time moves on it should give increasing information on default variability/predictability. If you could show which risk model was being used for each cohort so much the better.

[myles]

I've updated the summary document with a few more charts: summary.pdf

Would appreciate thoughts on the last chart - Estimated vs Actual Default Rate - which I've tidied up and made more informative (I hope). Harmoney, in their definition of Estimated Default Rate, make no mention of limiting the 'window' of loans by ignoring the first period where loans don't default/aren't flagged as defaults. This is the approach I've taken to be consistent with their figures (i.e. no window).

However, other platforms, e.g. Lending Club, limit the window to loans older than 120 days:

"Annualized Charge Off Rate is calculated by dividing the total amount of loans in charge off by the total amount of loans issued for more than 120 days, divided by the number of months loans in charge off have been outstanding and multiplied by twelve. The loans issued for less than 120 days are excluded from the calculation because loans are unlikely to charge off during the first 120 days."
Source: Lending Club - How We Measure Annualized Charge Off Rate

If anyone has thoughts on this, I'd appreciate them.

[beacon]

Thoughts on this style:

Attachment 10063

Great graph, Myles. Illuminating to see defaults reaching a quarter of the loans already, in the earlier cohorts for DEF.

[BJ1]

Capture131018.JPGHow many willing buyers are there for this risk? I bet it gets filled and defaults - but none of my money is going to repayments ratios like this..

[beacon]

Harmoney ... make no mention of limiting the 'window' of loans by ignoring the first period where loans don't default/aren't flagged as defaults. This is the approach I've taken to be consistent with their figures (i.e. no window). ... Lending Crowed, limit the window to loans older than 120 days:

"Annualized Charge Off Rate is calculated by dividing the total amount of loans in charge off by the total amount of loans issued for more than 120 days, divided by the number of months loans in charge off have been outstanding and multiplied by twelve. The loans issued for less than 120 days are excluded from the calculation because loans are unlikely to charge off during the first 120 days."
Source: Lending Crowd - How We Measure Annualized Charge Off Rate

If anyone has thoughts on this, I'd appreciate them.

I agree that using the all loans approach is valuable for comparative purposes (with Harmoney) and easier, but Lending Club approach of excluding the <120 day laons and annualizing the rest is more technically correct. I would have used the former.

[beacon]

... there's significantly more AB grades in home improvement and 50-59 age group loans; and more EF grades in 20-29yo and household items (for example).

The question begs whether a "C" grade 20-29yo is more likely to default compared to a "C" grade 50-59, and similar with Home Improvements vs HH Items or Used Cars. If not those categories may already be fully explained by the grading category assigned by HM.

Commend you on your efforts so far though, top work!

Attachment 10064

I had observed in my data too that riskier grades were predominantly youngsters and safer grades elders, so there is an element of safety with age already built into HM grading, I think. But interesting insights on household items vs home improvements, leesal.

[beacon]

I've updated the summary document with a few more charts: summary.pdf

Would appreciate thoughts on the last chart - Estimated vs Actual Default Rate - which I've tidied up and made more informative (I hope).

Thanks Myles, I still don't get what you are plotting here - and what it is showing? Also, the samples (bottom numbers on x-axis) in borrower profile loan characteristic charts don't seem to add up to 21492. Can you re check please?

[myles]

I still don't get what you are plotting here - and what it is showing?

The Orange line is the Harmoney suggested default rate - this is plotted against itself to produce the line. So one of the points that make up this line is 3.15% vs 3.15% - this is the Scorecard 1.5 estimated default rate for an E1 risk grade (details here). There is a similar point for each estimated default rate for all risk grades (Scorecard 1.0 and 1.5 - sourced from the data set) that makes up the line.

The scatter graph is plotting the recorded estimated default rate for each data set record (grouped by estimated default rate) against what the actual default rate for that group is. This puts a point on the graph at x=Estimated Default Rate vs y=Actual Default Rate. The circle is proportional to the size of the population of loans with that estimated default rate.

Hmm, not sure how clear that is...

Basically showing what Harmoney say the default is estimated to be vs what it actually is (grouped by risk grade).............

[myles]

Also, the samples (bottom numbers on x-axis) in borrower profile loan characteristic charts don't seem to add up to 21492. Can you re check please?

Shortfall will be Cancelled loans ... a quick check shows them adding up to 21308? Database count confirms 21308 as loans not Cancelled.

Summary at top of report may not be completely correct - will be when I finalise the data set.

[myles]

Just so there is a comparison of re-write vs non re-write default rates (based on my own loan set only, as this detail can't be captured from the csv):

pd_re-write.jpg