Cumulative Gross Loss Rates By Vintage
Navigate to Analysis in Performance
X-Axis = Loan Age
Left Y-Axis = Cum Charge Off
Select Base Performance Line = Cum Charge Off (shown in the first graph)
Strat By = Vintage-Quarter
Filter = Vintage-Quarter for 2022 - 2023
This allows you to understand an apples to apples comparison across different vintages from a gross loss perspective, not accounting for recoveries. We also have a line showing the aggregate weighted average for gross losses.
We can get more granular by using Chart By FICO-Original to understand how credit quality affects this performance metric. In this case, we can see higher losses are attributed to lower FICO bands.
Prepayment Rates By GWAC
Navigate to Analysis in Performance
Left Y-Axis = CPR
Strat By = GWAC
Select Cog icon named Chart Settings
Select Edit Buckets so you can amend the numeric groupings for GWAC
Linear Bucket - input the Range From, Range To, and Increment Value (10 to 25 by 5pt increments)
Advanced Bucket - allows non-linear groupings, for example 50pt increments from 500 to 650 and 25pt increments from 650 to 800 (500 to 650 by 50; 650 to 800 by 25)
This allows you to understand how prepayments can be affected by GWACs bands.
We can get more granular by using Strat By GWAC and Chart By FICO-Original to understand how credit quality affects CPRs. In this case, we can see higher FICOs have higher prepayment rates in each GWAC band, which could be a result of more opportunities to refinance/prepay due to borrowers' financial situation.
State Concentration Through Different Perspectives
Option 1 - Generic Strat
Navigate to Reports tab
Strat by State
Filter if desired by a state through selecting state filters on tables or right clicking on a state
Option 2 - Custom Columns
Navigate to Reports tab
Select Edit Columns (drawer icon) o right side and Create New Column
Customize the name of the field
Function = Percent of Bucket
Target Column = Original Balance
Filter Formula = {Borrower State at Origination} =”MA”
Add or move this newly built column to where you desire in your table
Option 3 - Geo Map
Navigate to Geo Map
Geography Level = State
Loan Attribute = count
Can choose from variety of metrics, whether it is count, performance focused (losses, delinquencies, etc.)
Helpful in understanding how macro factors or natural disasters could affect geographical areas.
Leveraging Document Type on dv01
On dv01, we have normalized over 400+ values that platforms and servicers have provided to us in the mortgage sector to 15 different categories that can be seen in our Non-QM and Jumbo benchmarks as well as securitizations and whole loan portfolios. This allows for simplicity and transparency among a popular attribute used for credit analysis across different datasets in this sector for our investor base and issuers on dv01. The below shows how you can use this attribute for our Non-QM benchmark:
Navigate to Reports page
Strat by Doc Type
From here, you can incorporate filters on doc type values like DSCR or certain bank statements, which is helpful if an investor owns loans with a particular doc type or if a lender underwrites to certain criteria.
Additionally, we can move away from static analysis and incorporate time series analysis using the following:
Navigate to Analysis in Performance
Y-Axis = % DQ
% DQ BY = Percent 60+ DQ
Strat By = Doc Type
We can leverage custom groupings if you wanted to compare categories against one another, which is popular for DSCR vs. Non-DSCR loans. The process is the following:
Navigate to Chart Settings and select Edit Buckets
Toggle over to Numeric Groupings and from the dropdown select the doc type field, check the box on anything other than DSCR, and name the grouping as “Other”
Loan Rate Incentives (S-Curves)
Navigate to Analysis in Performance
Toggle the X-Axis to Current Loan Incentive
Rate incentive between the current note rate and PMMS rate
Allows you to examine sensitivity of metrics and composition of collateral to changes in interest rates
Amend the Date Range to 2020-01 - 2024-06
Filter the Vintage - Quarter for 2018+ vintages
Add CPR and % Curr Bal in the Left Y-Axis
Allows you to see how loans with the larger rent incentive tend to have higher CPRs at a certain breakage point. This is helpful for investors who have a portfolio of loans or CUSIPs where they are trying to think through price and prepayment assumptions based on the underlying loans' coupon and PMMS rate.
Compare Performance Across Datasets
Users have the ability to compare multiple datasets with an asset class against each other. This is powerful for investors who are looking to evaluate performance or composition on their portfolio, platforms, and/or securitizations they are invested against the datasets they have access to on dv01. In addition, it is also helpful for investors or issuers looking to amend their underwriting criteria, competitiveness, and the assumptions they should use for modeling.
Popular metrics to compare against could be cumulative gross loss, delinquencies, CPRs, CDRs, and understanding weighted average metrics over time.
Navigate to Analysis in Performance
Select two or more datasets
Fitch-dv01 Prime Jumbo Benchmark
Fitch-dv01 Non-QM Benchmark
Left Y-Axis = Curr GWAC
Strat By = Pool
Can Filter by multiple attributes
Borrower Original FICO = 600 to 700
Original Combined LTV = 0 to 90
Loan ID Search (Exact and Partial)
Navigate to Loan Tape if viewing a securitization or portfolio
Search all ID fields for a string via the search bar on the left
Exact ID - Search 7bb2c2-c008-4af1-aa0e-adf100089d1c in the search bar returns the loan match and clicking the blue ID, opens the LoanID Card to see granular loan-level characteristics.
Partial ID - Search 1000 returns all IDs that contain 53 across all ID fields
