Extended Viewer

Joy Zhang

Joy Zhang
Executive Director, MSCI Research

About the Contributor

Joy Zhang is an Executive Director and Head of Non-Agency Securitization Research. Previously, Joy was a Director at Credit Suisse, responsible for mortgage collateral and regulatory modeling for securitized products trading. She also has worked as a senior developer at Goldman Sachs responsible for developing a firm-wide risk management system. Joy has an M.S. in Computational Finance from the Carnegie Mellon University and a Ph.D. in Chemistry from University of Kansas.

HTML Displayer Portlet

Contributions by Joy Zhang


Nothing was found.
  1. BLOG

    COVID Stimulus Helped Resilience of US ABS 

    Jan 29, 2021 Yini Yang , Joy Zhang

    Risk Management , Fixed Income

    Learn More

    Issuance of U.S. asset-backed securities fell by a quarter in 2020 from the previous year, as credit tightened during the COVID-19 crisis. The performance of loans underpinning ABS proved resilient, however, as economic relief helped support consumers.

  2. PAPER

    Sensitivity-based MBS VaR methodology 

    Sep 28, 2020 Joy Zhang

    Factor and Risk Modeling , Risk Management

    Download Document

    In this document, we evaluate the methodology for MBS VaR using sensitivity approximation.

  3. BLOG

    Are Securitized Products Ready for the LIBOR-SOFR Transition? 

    Sep 2, 2020 Joy Zhang , Yihai Yu

    Fixed Income , Integrated Risk Management

    Learn More

    Will the securitization industry be ready for the transition from LIBOR to the secured overnight financing rate (SOFR), as it faces the fact that LIBOR can no longer be guaranteed beyond the end of 2021? As the industry mobilizes, significant challenges remain.

  4. BLOG

    Consumer ABS: Recovering from Coronavirus? 

    Jun 11, 2020 Yini Yang , Jian Chen , Joy Zhang

    Risk Management , Fixed Income

    Learn More

    After the U.S. COVID-19 lockdown, new monthly remittance reports for asset-backed securities indicated performance deterioration and signaled potential challenges ahead. Meanwhile, in China, ABS showed signs of recovery.

  5. BLOG

    Can AI Model the Complexities of MBS Prepayment? 

    May 29, 2020 Joy Zhang , Yini Yang

    Fixed Income , Risk Management

    Learn More

    Machine learning using neural networks has been successfully applied to fields in which extremely complex patterns can prove challenging for other algorithms. Are neural networks suited for modeling prepayment risk in agency mortgage-backed securities?

  6. BLOG

    Consumer ABS Under Coronavirus in the US and China 

    May 11, 2020 Yini Yang , Jian Chen , Joy Zhang

    Emerging Markets , Fixed Income , Global Investing , Risk Management

    Learn More

    Beyond COVID-19’s steep human toll, the pandemic’s disruption of economic life has led to widespread loss of income and impaired some borrowers’ ability to repay loans. What could the impact be for investors in consumer asset-backed securities in the U.S. and China?

  7. BLOG

    Could coronavirus depress US housing prices? 

    Apr 15, 2020 Yihai Yu , Joy Zhang

    Risk Management , Fixed Income

    Learn More

    The large economic shocks unleashed by the coronavirus pandemic could be comparable to or even exceed those of the 2008 global financial crisis (GFC). We used our models to assess whether these shocks could hurt U.S. housing prices as much as the GFC did.

  8. BLOG

    Could coronavirus lead to default contagion in CLOs? 

    Apr 1, 2020 Yini Yang , Joy Zhang

    Risk Management , Fixed Income

    Learn More

    The market for collateralized loan obligations is under severe stress during the COVID-19 pandemic. We used MSCI’s loan and CLO models to assess a sample CLO’s loan-default risk characteristics. Could a wave of defaults harm CLOs?

  9. PAPER

    MSCI Agency Prepayment Model Performance Review 2019 

    Feb 10, 2020 Joy Zhang

    Factor and Risk Modeling , Risk Management

    Download Document

    As required by the MSCI Securitized Product Model policy, we review and approve MSCI agency fixed rate prepayment model performance in 2019. The review includes mainly three areas -Prepayment forecasts: The model gave accurate prepayment forecasts for major vintage coupon cohorts. The weighted average 6-month average forecasts errors are below 0.2 smm, with most of the major vintage coupon cohorts having an error below 0.3 smm. The model was able to differentiate prepayment behavior across collateral attributes and macro-economic environment, mainly the significant rates rally in the summer. The average ranked based error measure is 0.2 smm. -MBS TBA durations: The model was able to track empirical durations, hence the market price dynamics, across the significant rates rally. Bucketed based on TBA moneyness, the model durations were able to track empirical durations within 0.25 year. -VaR measure: The model OAS time series were found to be stationary, hence suitable for VaR measures. The KS statistic for the p-value of actual OAS change vs. forecasted distribution based on OAS history is less than 0.04. 2019 saw significant rates rallies and mortgage basis volatilities, as well as major uncertainties in Fed’s monetary policy. The model performance was also reviewed in the context of Press reports of wide spread model under-forecasts of the large prepayment during the summer refinance wave.1 As suggested by best practice, the appendix presents some model comparisons with three other models from peer analytics providers.

  10. Mortgage-backed securities make up a huge portion of the U.S. financial system. Mortgage prepayment modeling is essential in MBS investment and risk analysis. It is also among the most complex areas of financial modeling, because of the vast data volume and large number of risk factors. Does an AI-based model using neural networks have an advantage over traditional models in taking on the complexities of MBS prepayment risk?

  11. BLOG

    MBS prepayment modeling: AI 1, Humans 0? 

    Sep 27, 2019 David Zhang , Joy Zhang

    Fixed Income , Models/Client Cases , Risk Management

    Learn More

    Artificial intelligence has broken through in fields previously dominated by humans. Could AI surpass humans in modeling the complex risks of agency mortgage-backed securities?

  12. BLOG

    MBS investors: quantitative easing déjà vu? 

    Sep 5, 2019 Joy Zhang

    Fixed Income , Risk Management

    Learn More

    Despite the Fed’s silence on the matter, the MBS market may be indicating that a new round of QE is coming.

  13. PAPER

    MSCI Agency Fixed Rate Default Model: The Effect of Borrower Subsequent Debt 

    May 31, 2019 Yini Yang , Joy Zhang

    Download Document

  14. PAPER

    MSCI House Price Volatility Model: Spatial and Temporal Structure 

    May 30, 2019 Joy Zhang , Yini Yang

    Download Document

  15. BLOG

    Are Subprime Auto Loans at a Tipping Point? 

    Mar 4, 2019 Yini Yang , Joy Zhang

    Models/Client Cases , Economic Exposure , Fixed Income

    Learn More

    Investors and the media have lately turned their attention to credit risk in U.S. subprime automotive lending — concerns that increased during the recent market volatility.

  16. PAPER

    MSCI US Auto Loan Collateral Model Insight 

    Oct 29, 2018 Yini Yang , Joy Zhang

    Factor and Risk Modeling , Risk Management

    Download Document

    The MSCI U.S. Auto Loan ABS Collateral models are used to project collateral loan performance, including prepayment, default, and loss severity rates. These are used for measuring the risk and return of these securities.

  17. BLOG

    Is MBS refinance risk increasing? 

    Oct 12, 2018 Joy Zhang

    Integrated Risk Management

    Learn More

    With the Federal Reserve raising interest rates and the majority of agency mortgage-backed securities (MBS) under the refinance threshold, how much do investors need to worry about refinance risk? Our model indicates that future refinance regimes would be similar to recent 2016 experiences, and this view is consistent with current behavior of MBS empirical durations. However, investors may want to remain vigilant, as the recent trend toward looser mortgage credit standards by agencies and regulators could increase the prepayment intensity of future refinance waves.

  18. BLOG

    As credit risk rises, beware of selection bias in CLOs 

    Jun 1, 2018 Joy Zhang

    Fixed Income

    Learn More

    Investors in the booming U.S. Collateralized Loan Obligation (CLO) market likely need to be aware of the risks: record tight spreads, deteriorating credit quality, and, as our expanded CLO data analytics reveal, selection bias risk.