On a Method of Estimating Birth and Death Rates and the Extent of Registration

Estimating drowning deaths in Northern Iran using capture–recapture method

“…This simple technique was first applied in the 1940 [20] and become more widespread after the work of Witte in the 1970s [21] followed by Stephen (1996) [22].…”

Inferring the default rate in a population by comparing two incomplete default databases

“…Sekar and Deming (1949) show that if there is positive (negative) correlation between D 1 and D 2 then the estimate in Eq. (4) is asymptotically biased downward (upward).…”

“…We found that, depending on the level of p 1 and p 2 , for low numbers of defaults (typically 1000 or 2000, depending on the level of missingness) the tests suggested by Diebold et al (1998) tend to reject the appropriateness of the Sekar and Deming (1949) asymptotic estimate suggesting that the asymptotic assumptions do not hold in these sparser data contexts. …”

“…Following the notation of Sekar and Deming (1949) let C denote the number of defaults detected in both databases N 1 denote the number of defaults detected in database 1 but not database 2 N 2 denote the number of defaults detected in database 2 but not database 1 N denote the total number of defaults M 1 denote the number of defaults detected in database 1 (N 1 + C) M 2 denote the number of defaults detected in database 2 (N 2 + C) D 1 denote a random variable with 1 indicating that a given default is captured in database 1 and 0 otherwise D 2 denote a random variable with 1 indicating that a given default is captured in database 2 and 0 otherwise…”

“…In situations where a researcher has access to two incomplete default data sets, it is possible to infer the number of ''missing'' defaults using a procedure developed by Sekar and Deming (1949). 3 We demonstrate this approach in this short note and show an application to defaults for North American public firms with less than $1 mm in sales.…”