An Empirical Evaluation of Structural Credit Risk Models

Abstract

This paper evaluates the capacity of five structural creditrisk models to forecast default rates. In contrast to previous studies with similar objectives, the paper employs firm-level data and finds that model-based forecasts of default rates tend to be unbiased and to deliver point-in-time errors that are small in both statistical and economic terms. In addition, inand out-of-sample regression analysis reveals that the models account for a significant portion of the variability of credit risk over time but fail to fully reflect its dependence on macroeconomic cycles.JEL Codes: G33, E44, G28, C13.

Capital regulation, risk-taking and monetary policy: A missing link in the transmission mechanism?

“…the coexistence of unusually rapid expansion in credit and asset prices, have useful predictive content for subsequent widespread financial distress, output weakness and disinflation, over horizons of two to four years ahead, depending on the calibration Lowe, 2002, 2004). Moreover, such macro indicators can also help to improve the predictive content of popular micro models of default risk, including those widely used in the financial industry (Tarashev, 2005).…”

Probabilidad de incumplimiento de entidades financieras colombianas: una aproximación estructural

“…Los trabajos que evalúan empíricamente los distintos métodos son escasos y en general no abordan el problema con un enfoque estructural. Algunos ejemplos son Tarashev (2005), Switzer and Wang (2013), Altman et al (2010), Bharath and Shumway (2008) y Kato and Hagendorff (2010). En este sentido, en este documento se adapta el modelo estructural de Goldstein et al (2001) a los datos de balance y mercado de las firmas del sector financiero que cotizan en el mercado de valores de Colombia.…”

“…Dado que no se tiene una estimación de este valor en Colombia, se escogió siguiendo la aproximación propuesta enTarashev (2005).…”

Default Risk Characteristics of Poll-Based Bond Spreads

“…Similarly, Zhu (2008) observed that replacing the Gaussian distribution of asset return with three alternative non-Gaussian distributions signifi cantly improves their performance. Tarashev (2008) found that with fi rm-level data, both the exogenous and endogenous structural credit risk models tend to account for the level of default rates observed in the market.…”