bias. We study this by using a data set that combines survey information of individual workers with administrative records … survey respondents and non-respondents. We develop a method to distinguish empirically between two explanations for a bias in … both explanations. We discuss implications for standard methods to deal with non-response bias. …

generates a higher income, but only for entrepreneurs: This finding supports Lazear's Jack-of-all-Trades theory. …

high-performance fuzzy firm name matching algorithm that uses existing computational methods and works even under hardware … improvement is due to cleaning, and another 41% (50%) improvement is due to similarity matching. 18.1% of all patent applications …

explore what combination of the expenditure data and the income data seem to produce the most sensible data according to a few …

-generation process, the least-squares estimators have smaller bias (in fact zero bias) but larger variances in the long regression than … in the short regression. But if the long regression is also misspecified, the bias may not be smaller. We provide bias …