SUMMARY A linearization theorem represents an estimating equation for an unknown parameter as approximately linear in the neighbourhood of the true parameter point, and can provide a convenient approach to asymptotic theory. A multiparameter version of a standard linearization theorem is proved, and applied to two examples: M-estimates in multiple regression, and the spherical median.
No takes yet. Share an insight, caveat, or question.
B. M. Brown (1985) studied this question.