Maximum likelihood estimation of first-passage structural credit risk models correcting for the survivorship bias

We study the statistical properties of the maximum likelihood estimator (MLE) accounting for survivorship bias for models based on the first-passage of the geometric Brownian motion. We find that neglecting the survivorship... [ view full abstract ]

We study the statistical properties of the maximum likelihood estimator (MLE) accounting for survivorship bias for models based on the first-passage of the geometric Brownian motion. We find that neglecting the survivorship bias overestimates the drift while conditioning on survival underestimates it. Therefore, we propose a bias correction method for non-iid samples that is first-order unbiased and second-order efficient. With an empirical study based upon a sample of more than 13,000 companies over the period 1980 through 2016 inclusive, our results show the economic significance of over- or under-correcting for survivorship.