Two-step estimation of linear models with ordinal unobserved variables: The case of corporate bonds

Chihwa Kao, Chunchi Wu

Research output: Contribution to journalArticlepeer-review

25 Scopus citations

Abstract

This article proposes a two-step method for estimating the impact of bond indenture provisions and other financial variables on the risk and yields of investment-grade and speculative corporate bonds. In the first step, the default risk of bonds is estimated as a function of indenture provisions and the characteristics of bonds and the issuing firms by an ordered probit. In the second step, the effects of default risk and bond characteristics on yields are estimated after a measure of bond default risk is obtained by a conditional-mean method.

Original languageEnglish (US)
Pages (from-to)317-325
Number of pages9
JournalJournal of Business and Economic Statistics
Volume8
Issue number3
DOIs
StatePublished - Jul 1990

Keywords

  • Conditional expectation
  • Default risk
  • Ordered probit

ASJC Scopus subject areas

  • Statistics and Probability
  • Social Sciences (miscellaneous)
  • Economics and Econometrics
  • Statistics, Probability and Uncertainty

Fingerprint

Dive into the research topics of 'Two-step estimation of linear models with ordinal unobserved variables: The case of corporate bonds'. Together they form a unique fingerprint.

Cite this