scientific article; zbMATH DE number 4001211

From MaRDI portal
Publication:4727205





Suppose \(x_ 1,...,x_ n\) are i.i.d. r.v.'s with distribution F. Let \(F_ n\) be the empirical distribution, and \(x^*_ 1,...,x^*_ n\) an i.i.d. sample from \(F_ n\). Let \(R(x_ 1,...,x_ n;F)\) be a random variable of interest. In the bootstrap literature, \(R(x^*_ 1,...,x^*_ n;F_ n)\) is adopted as an approximation to \(R(x_ 1,...,x_ n;F)\). If both \(R(x^*_ 1,...,x^*_ n;F_ n)\) and \(R(x_ 1,...,x_ n;F)\) have the same limit distribution, we say that the bootstrap approximation holds in this case. The paper shows that the bootstrap approximation for the M-estimate of the unknown coefficients in linear regression models holds under a set of conditions.











This page was built for publication:

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4727205)