A Conditionally Beta Distributed Time-Series Model With Application to Monthly US Corporate Default Rates
Publikation: Working paper › Forskning
Standard
A Conditionally Beta Distributed Time-Series Model With Application to Monthly US Corporate Default Rates. / Nielsen, Thor Pajhede.
2017.Publikation: Working paper › Forskning
Harvard
APA
Vancouver
Author
Bibtex
}
RIS
TY - UNPB
T1 - A Conditionally Beta Distributed Time-Series Model With Application to Monthly US Corporate Default Rates
AU - Nielsen, Thor Pajhede
PY - 2017
Y1 - 2017
N2 - We consider an observation driven, conditionally Beta distributed model for variables restricted to theunit interval. The model includes both explanatory variables and autoregressive dependence in the meanand precision parameters using the mean-precision parametrization of the beta distribution suggested byFerrari and Cribari-Neto (2004). Our model is a generalization of the βARMA model proposed in Rocha andCribari-Neto (2009), which we generalize to allow for covariates and a ARCH type structure in the precisionparameter. We also highlight some errors in their derivations of the score and information which has implicationsfor the asymptotic theory. Included simulations suggests that standard asymptotics for estimators andtest statistics apply. In an empirical application to Moody’s monthly US 12-month issuer default rates in theperiod 1972 − 2015, we revisit the results of Agosto et al. (2016) in examining the conditional independencehypothesis of Lando and Nielsen (2010). Empirically we find that; (1) the current default rate influence thedefault rate of the following periods even when conditioning on explanatory variables. (2) The 12 monthlag is highly significant in explaining the monthly default rate. (3) There is evidence for volatility clusteringbeyond what is accounted for by the inherent mean-precision relationship of the Beta distribution in thedefault rate data.
AB - We consider an observation driven, conditionally Beta distributed model for variables restricted to theunit interval. The model includes both explanatory variables and autoregressive dependence in the meanand precision parameters using the mean-precision parametrization of the beta distribution suggested byFerrari and Cribari-Neto (2004). Our model is a generalization of the βARMA model proposed in Rocha andCribari-Neto (2009), which we generalize to allow for covariates and a ARCH type structure in the precisionparameter. We also highlight some errors in their derivations of the score and information which has implicationsfor the asymptotic theory. Included simulations suggests that standard asymptotics for estimators andtest statistics apply. In an empirical application to Moody’s monthly US 12-month issuer default rates in theperiod 1972 − 2015, we revisit the results of Agosto et al. (2016) in examining the conditional independencehypothesis of Lando and Nielsen (2010). Empirically we find that; (1) the current default rate influence thedefault rate of the following periods even when conditioning on explanatory variables. (2) The 12 monthlag is highly significant in explaining the monthly default rate. (3) There is evidence for volatility clusteringbeyond what is accounted for by the inherent mean-precision relationship of the Beta distribution in thedefault rate data.
KW - Faculty of Social Sciences
KW - Beta regression
KW - credit risk default rates
KW - contagion
KW - C12
KW - C22
KW - C32
KW - C50
M3 - Working paper
T3 - University of Copenhagen. Institute of Economics. Discussion Papers (Online)
BT - A Conditionally Beta Distributed Time-Series Model With Application to Monthly US Corporate Default Rates
ER -
ID: 178281805