Confidence Sets for Continuous-time Rating Transition Probabilities

Jens Henrik Eggert Christensen, Ernst Hansen, David Lando

Research output: Working paperResearch

9 Downloads (Pure)

Abstract

This paper addresses the estimation of default probabilities and associated
confidence sets with special focus on rare events. Research on rating transition data has documented a tendency for recently downgraded issuers to
be at an increased risk of experiencing further downgrades compared to issuers that have held the same rating for a longer period of time. To capture
this non-Markov effect we introduce a continuous-time hidden Markov chain
model in which downgrades firms enter into a hidden, ’excited’ state. Using
data from Moody’s we estimate the parameters of the model, and conclude
that both default probabilities and confidence sets are strongly influenced by
the introduction of hidden excited states.
Original languageEnglish
Number of pages37
Publication statusPublished - 2004

Cite this

Christensen, Jens Henrik Eggert ; Hansen, Ernst ; Lando, David. / Confidence Sets for Continuous-time Rating Transition Probabilities. 2004.
@techreport{5ebdd9b860154ec5a4256d874a946e52,
title = "Confidence Sets for Continuous-time Rating Transition Probabilities",
abstract = "This paper addresses the estimation of default probabilities and associatedconfidence sets with special focus on rare events. Research on rating transition data has documented a tendency for recently downgraded issuers tobe at an increased risk of experiencing further downgrades compared to issuers that have held the same rating for a longer period of time. To capturethis non-Markov effect we introduce a continuous-time hidden Markov chainmodel in which downgrades firms enter into a hidden, ’excited’ state. Usingdata from Moody’s we estimate the parameters of the model, and concludethat both default probabilities and confidence sets are strongly influenced bythe introduction of hidden excited states.",
author = "Christensen, {Jens Henrik Eggert} and Ernst Hansen and David Lando",
year = "2004",
language = "English",
type = "WorkingPaper",

}

Confidence Sets for Continuous-time Rating Transition Probabilities. / Christensen, Jens Henrik Eggert; Hansen, Ernst; Lando, David.

2004.

Research output: Working paperResearch

TY - UNPB

T1 - Confidence Sets for Continuous-time Rating Transition Probabilities

AU - Christensen, Jens Henrik Eggert

AU - Hansen, Ernst

AU - Lando, David

PY - 2004

Y1 - 2004

N2 - This paper addresses the estimation of default probabilities and associatedconfidence sets with special focus on rare events. Research on rating transition data has documented a tendency for recently downgraded issuers tobe at an increased risk of experiencing further downgrades compared to issuers that have held the same rating for a longer period of time. To capturethis non-Markov effect we introduce a continuous-time hidden Markov chainmodel in which downgrades firms enter into a hidden, ’excited’ state. Usingdata from Moody’s we estimate the parameters of the model, and concludethat both default probabilities and confidence sets are strongly influenced bythe introduction of hidden excited states.

AB - This paper addresses the estimation of default probabilities and associatedconfidence sets with special focus on rare events. Research on rating transition data has documented a tendency for recently downgraded issuers tobe at an increased risk of experiencing further downgrades compared to issuers that have held the same rating for a longer period of time. To capturethis non-Markov effect we introduce a continuous-time hidden Markov chainmodel in which downgrades firms enter into a hidden, ’excited’ state. Usingdata from Moody’s we estimate the parameters of the model, and concludethat both default probabilities and confidence sets are strongly influenced bythe introduction of hidden excited states.

M3 - Working paper

BT - Confidence Sets for Continuous-time Rating Transition Probabilities

ER -