Diligence on the loan tape, as the seller sent it.
Upload the tape in whatever shape it arrives. portfolio-dd maps it onto one standard schema, runs the strats, charge-off by segment and vintage curves you already run, and shows which segments sit outside the book. The sample is Lending Club’s book: 886,837 loans originated January 2008 to December 2015.
How it works
-
Ingest
Upload the tape as CSV, TSV, Excel or Parquet. No reshaping by hand first; the seller’s column names stay as they are.
-
Standardise
Each column is matched to the standard schema. A local model proposes matches for names it does not recognise, with a confidence and a reason, and you confirm every one before anything is cast.
-
Analyse
Strats, charge-off by segment, vintage curves, yield against loss, concentration and data quality, then the segments that sit outside the book and why.
The sample book
Loans originated January 2008 to December 2015, with status as of March 2019.
- Loans
- 886,837
- Original balance
- $13.1BSum of original balance.
- Weighted average coupon
- 13.6%Interest rate weighted by original balance.
- Charge-off rate (count)
- 17.2%Loans charged off at the snapshot over all loans.
Every figure on this site is read from the sample report when the page renders. All 8 headline figures, and how each is built.
What you get
- Stratifications by grade, term, purpose and geography: count, balance, share of book and WAC.
- Charge-off by segment against the book, so a weak segment shows up as a gap, not a number in isolation.
- Vintage curves cumulative charge-off by months on book, one line per grade.
- Yield against loss per grade, to see where pricing covers the losses and where it does not.
- Concentration in the largest states and purposes, as a share of balance.
- Data quality missing and out-of-range values per field, checked before any number is trusted.
- Outliers the segments that sit furthest from the book, with a plain-language reason for each.