Linear regression with randomly double-truncated data

Non-parametric estimation for a linear regression model under random double-truncation is investigated, i.e. the variables are observed if and only dependent variable lies in interval. The method requires weak distribution assumptions to ensure identifiability, but does not require any specific family variable, neither truncation nor error term. By using non-parametric estimators of several functions, consistent asymptotically normal established. A simulation study shows tendency that lower probability observation, higher mean squared estimators, even same number observations. Finally, applied doubly truncated data set German companies, where age-at-insolvency interest. Keywords: Insolvency risk, Linear regression, Non-parametric, Random