In agency pools, loans with balances below $200,000 offer prepayment protection (i.e., they prepay more slowly) relative to loans with higher balances. Servicers typically segregate these loans into specified pools that trade at a premium over TBA-deliverable pools. But the prepayment protection isn’t homogenous and varies significantly by state.
During last week's SFIG Residential Mortgage Finance Symposium, I moderated a panel on best practices in disclosure and reporting data related to private-label mortgage securities. We discussed many of the challenges confronting issuers, investors, rating agencies, and the industry with sharing relevant data in general and with implementing the SEC's Regulation AB II requirements in particular. Five minutes after my panel ended, my colleague Suhrud Dagli moderated a panel that discussed the applicability of blockchain technology to the securitization industry. Walking out of the symposium a short time later, I began to wonder how interesting it would have been if our two sessions had been combined.
In light of recent news stories[i] concerning efforts to stem aggressive solicitations that steer VA borrowers toward refinancings that are not necessarily in their best interest, we thought it fitting to take a look at some of the data underlying this trend. At issue are claims that VA borrowers are being persuaded to refinance their mortgages ostensibly to reduce their monthly payment. In many cases, however, the lower monthly payment was being made possible primarily by upfront fees and by extending the term of the mortgage. Consequently, even though the monthly mortgage payment was going down, the mortgage balance was often going up along with the number of payments required to pay off the mortgage. Let’s see what the data indicates.
Tuning is the process of maximizing a model’s performance without overfitting or creating too high of a variance. In machine learning, this is accomplished by selecting appropriate “hyperparameters.”
Anti-money-laundering (AML) solutions have no business being classified as models. To be sure, AML “models” are sophisticated, complex, and vitally important. But it requires a rather expansive interpretation of the OCC/Federal Reserve/FDIC<sup>1</sup> definition of the term <i>model</i> to realistically apply the term to AML solutions.
Machine learning model selection is the second step of the machine learning process, following variable selection and data cleansing. Selecting the right machine learning model is a critical step, as a model which does not appropriately fit the data will yield inaccurate results.
An accounting manager at a mid-sized bank recently wondered aloud to us how to approach implementing end-user computing (EUC) controls. She had recently become responsible for identifying and overseeing her institution's unknown number of EUC applications and had obviously given a lot of thought to the types of applications that needed to be identified and what the review process ought to look like. She recognized that a comprehensive inventory would need to be built, but, like so many others in her position, was uncertain of how to go about it.
Model evaluation is the process of objectively measuring how well machine learning models perform the specific tasks they were designed to do—such as predicting a stock price or appropriately flagging credit card transactions as fraud. Because each machine learning model is unique, optimal methods of evaluation vary depending on whether the model in question is “supervised” or “unsupervised.” Supervised machine learning models make specific predictions or classifications based on labeled training data, while unsupervised machine learning models seek to cluster or otherwise find patterns in unlabeled data.
With the upcoming implementation of IFRS 9 in 2018, the discussion of Basel III capital requirements is gaining importance. The relationship between capital and provisions for loan-loss has been a topic of discussion as the world moves towards mandating loss provisioning by looking out over the life of a financial asset. How will this new credit-loss approach for provisioning affect regulatory capital?
Feature selection in machine learning refers to the process of isolating only those variables (or “features”) in a dataset that are pertinent to the analysis. Failure to do this effectively has many drawbacks, including: 1) unnecessarily complex models with difficult-to-interpret outcomes, 2) longer computing time, and 3) collinearity and overfitting. Effective feature selection eliminates redundant variables and keeps only the best subset of predictors in the model, thus making it possible to represent the data in the simplest way.
With much anticipation and fanfare, the Federal Reserve is finally on track to reduce its MBS holdings. Guidance from the September FOMC meeting reveals that the Fed will allow its MBS holdings to “run off,” reducing its position via prepayments as opposed to selling it off. What does this mean for the market? In the long-term, it means a large increase in net supply of Agency MBS and with it an increase in overall implied and realized volatility.
Interestingly, the shape of a deal’s S-curve tends to vary depending on who is servicing the deal. Many things contribute to this difference, including how actively servicers market refinance opportunities. How important is it to be able to evaluate and analyze the S-curves for the servicers specific to a given deal? It depends, but it could be imperative.
Implementing CECL has brought about a host of accounting and other technical questions. The Financial Accounting Standards Board (FASB) works with the industry through a series of meetings to identify these questions, evaluate industry feedback, and periodically issue clarifying statements. We will continuously publish summarized points of interest from these meetings as they arise.
Weaknesses in securitization processes for mortgage loans contributed to the financial crisis of 2007 – 2008 and have led to a decade-long stagnation in the private-label residential mortgage-backed securities (PLS) market.
Although market participants have attempted to improve known weaknesses, lack of demand for private-label RMBS reflects investors’ reluctance to re-enter the market and the need for continued improvements to securitization processes to re-establish market activity. While significant issues still need to be addressed, promising advances have been made in the PLS market that improve information provided to investors as well as checks and balances designed to boost transaction performance.
Last year the Federal Housing Finance Agency (FHFA)—Fannie Mae’s and Freddie Mac’s regulator—announced a streamlined version of the federal government’s popular Home Affordable Refinance Program (HARP). The streamlined program will expand HARP eligibility to include mortgages originated on or after October 1, 2017.
Validating short-rate models can be challenging because many different ways of modeling how interest rates change over time (“interest rate dynamics”) have been created over the years. Each approach has advantages and shortcomings, and it is critical to distinguish the limitations and advantages of each of them to understand whether the short-rate model being used is appropriate to the task. This can be accomplished via the basic tenets of model validation—evaluation of conceptual soundness, replication, benchmarking, and outcomes analysis. Applying these concepts to short-rate models, however, poses some unique complications.
There are literally thousands of different servicers of Fannie Mae and Freddie Mac pools. Does the servicer make any material difference in the prepayment speed?
Increasing regulatory scrutiny due to the catastrophic risk associated with anti-money-laundering (AML) non-compliance is prompting many banks to tighten up their approach to AML model validation. Because AML applications would be better classified as highly specialized, complex systems of algorithms and business rules than as “models,” applying model validation techniques to them presents some unique challenges that make documentation especially important.
Attribution analysis of portfolios typically aims to discover the impact that a portfolio manager’s investment choices and strategies had on overall profitability. They can help determine whether success was the result of an educated choice or simply good luck. Usually a benchmark is chosen and the portfolio’s performance is assessed relative to it.
This post, however, considers the question of whether a non-referential assessment is possible. That is, can we deconstruct and assess a portfolio’s performance without employing a benchmark? Such an analysis would require access to historical return as well as the portfolio’s weights and perhaps the volatility of interest rates, if some of the components exhibit a dependence on them. This list of required variables is by no means exhaustive.
An instrument’s terms and conditions lie at the heart of cash flow generation and valuation. Not surprisingly, errors in terms and conditions can drive errors in valuation. Fortunately, fixing these errors is often straightforward, provided the terms and conditions data is readily available, which is not always the case for private placement instruments.