The predictive strength of MBS yield spreads during asset bubbles

Solomon Deku; Alper Kara; Artur Semeyutin

doi:10.1007/s11156-020-00888-8

The predictive strength of MBS yield spreads during asset bubbles

Solomon Deku, Alper Kara, Artur Semeyutin

School of Economics, Finance and Accounting

Research output: Contribution to journal › Article › peer-review

1 Citation (Scopus)

4 Downloads (Pure)

Abstract

We examine whether the predictive power of initial yield spreads of mortgage-backed securities (MBS) vary with the financial cycle. Using a cross-country sample of 4203 MBS, we find that initial yield spreads of MBS incorporate more information than credit ratings and predict future downgrades, even after conditioning on initial credit ratings. Predictive power of spreads is higher during credit and housing bubbles and for the least risky AAA-rated MBS. We find that initial yield spreads capture the magnitude of rating downgrades, especially during asset bubble periods. As a novel approach in this literature, we also utilise machine learning techniques (regression trees, naïve Bayes, support vector machines and random forests) to confirm our results.

Original language	English
Pages (from-to)	111-142
Number of pages	32
Journal	Review of Quantitative Finance and Accounting
Volume	56
Issue number	1
Early online date	29 Apr 2020
DOIs	https://doi.org/10.1007/s11156-020-00888-8
Publication status	Published - Jan 2021

Bibliographical note

This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material.

Funder

The authors thank David Marques-Ibanez, Philip Molyneux, Yener Altunbas, Alberto Pozzolo and Mark Rhodes for helpful comments and discussions. Our thanks also to participants at the 2018 Financial Data Science and Econometrics Workshop at the Loughborough University, the 2017 Southern Finance Association?s Key West Conference, the 4th European Conference on Banking and the Economy, and at seminars held at the University of Huddersfield and Nottingham Trent University.

Keywords

Asset bubbles
Credit ratings
MBS pricing
Machine learning
Securitization

ASJC Scopus subject areas

Accounting
Business, Management and Accounting(all)
Finance

Access to Document

10.1007/s11156-020-00888-8Licence: CC BY

PublishedFinal published version, 2.04 MBLicence: CC BY

Cite this

@article{bb6e66a1447542b28e486f26e20f01d6,

title = "The predictive strength of MBS yield spreads during asset bubbles",

abstract = "We examine whether the predictive power of initial yield spreads of mortgage-backed securities (MBS) vary with the financial cycle. Using a cross-country sample of 4203 MBS, we find that initial yield spreads of MBS incorporate more information than credit ratings and predict future downgrades, even after conditioning on initial credit ratings. Predictive power of spreads is higher during credit and housing bubbles and for the least risky AAA-rated MBS. We find that initial yield spreads capture the magnitude of rating downgrades, especially during asset bubble periods. As a novel approach in this literature, we also utilise machine learning techniques (regression trees, na{\"i}ve Bayes, support vector machines and random forests) to confirm our results.",

keywords = "Asset bubbles, Credit ratings, MBS pricing, Machine learning, Securitization",

author = "Solomon Deku and Alper Kara and Artur Semeyutin",

note = "This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. ",

year = "2021",

month = jan,

doi = "10.1007/s11156-020-00888-8",

language = "English",

volume = "56",

pages = "111--142",

journal = "Review of Quantitative Finance and Accounting",

issn = "0924-865X",

publisher = "Springer",

number = "1",

}

TY - JOUR

T1 - The predictive strength of MBS yield spreads during asset bubbles

AU - Deku, Solomon

AU - Kara, Alper

AU - Semeyutin, Artur

N1 - This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material.

PY - 2021/1

Y1 - 2021/1

N2 - We examine whether the predictive power of initial yield spreads of mortgage-backed securities (MBS) vary with the financial cycle. Using a cross-country sample of 4203 MBS, we find that initial yield spreads of MBS incorporate more information than credit ratings and predict future downgrades, even after conditioning on initial credit ratings. Predictive power of spreads is higher during credit and housing bubbles and for the least risky AAA-rated MBS. We find that initial yield spreads capture the magnitude of rating downgrades, especially during asset bubble periods. As a novel approach in this literature, we also utilise machine learning techniques (regression trees, naïve Bayes, support vector machines and random forests) to confirm our results.

AB - We examine whether the predictive power of initial yield spreads of mortgage-backed securities (MBS) vary with the financial cycle. Using a cross-country sample of 4203 MBS, we find that initial yield spreads of MBS incorporate more information than credit ratings and predict future downgrades, even after conditioning on initial credit ratings. Predictive power of spreads is higher during credit and housing bubbles and for the least risky AAA-rated MBS. We find that initial yield spreads capture the magnitude of rating downgrades, especially during asset bubble periods. As a novel approach in this literature, we also utilise machine learning techniques (regression trees, naïve Bayes, support vector machines and random forests) to confirm our results.

KW - Asset bubbles

KW - Credit ratings

KW - MBS pricing

KW - Machine learning

KW - Securitization

UR - http://www.scopus.com/inward/record.url?scp=85084224841&partnerID=8YFLogxK

U2 - 10.1007/s11156-020-00888-8

DO - 10.1007/s11156-020-00888-8

M3 - Article

VL - 56

SP - 111

EP - 142

JO - Review of Quantitative Finance and Accounting

JF - Review of Quantitative Finance and Accounting

SN - 0924-865X

IS - 1

ER -

The predictive strength of MBS yield spreads during asset bubbles

Abstract

Bibliographical note

Funder

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this